# Water in the upper mantle and deep crust of eastern China: concentration, distribution and implications

Water in the upper mantle and deep crust of eastern China: concentration, distribution and... Abstract Understanding the concentration and distribution of water in the Earth's mantle plays a substantial role in studying its chemical, physical and dynamic processes. After a decade of research, a comprehensive dataset of water content in upper-mantle samples has been built for eastern China, which is now the only place with water-content data from such diverse types of natural samples, and provides an integrated picture of the water content and its distribution in the upper mantle at a continental scale. The main findings include the following: (i) the temporal heterogeneity of the water content in the lithospheric mantle from early Cretaceous (∼120 Ma) to Cenozoic (<40 Ma) was tightly connected with the stability of the North China Craton (from its destruction to its consolidation); (ii) the heterogeneous water content in the Cenozoic lithospheric mantle beneath different blocks of eastern China was not only inherited from tectonic settings from which they came, but was also affected later by geological processes they experienced; (iii) the distinct water content between the lowermost crust and lithospheric mantle of eastern China and its induced rheological contrast at the base of the crust indicate that the continental crust–mantle boundary could behave either in a coupled or decoupled manner beneath different areas and/or at different stages; (iv) the alkali basalts of eastern China demonstrate a heterogeneous distribution of water content in the mantle; local and regional comparisons of the water content between the lithospheric mantle and basalts’ source indicate that the Cenozoic alkali basalts in eastern China were not sourced from the lithospheric mantle. Instead, the inferred high water contents in the mantle sources suggest that the Cenozoic eastern China basalts were likely sourced from the mantle transition zone (MTZ); and (v) both oceanic and continental crusts may carry a certain amount of water back into the deep mantle of eastern China by plate subduction. Such recycled crustal materials have not only created a local water-rich zone, but have also introduced crustal geochemical signatures into the mantle, both accounting for crustal geochemical imprints in the intra-plate magmatic rocks of eastern China. water, upper mantle and deep crust, continental stability, basalt genesis, eastern China INTRODUCTION Hydrogen that is structurally bound to other ions (mainly oxygen) in minerals is traditionally referred to as ‘water’ in the Earth science community and its concentration is calculated as H2O by weight. Its existence, even at trace levels (ppm level), can significantly influence certain important chemical and physical properties (e.g. seismic velocities, electrical and thermal conductivities, rheology, optical properties, melting temperature, pressure and phase relations and ion diffusion rate) of minerals and therefore rocks [1–13]. Consequently, water affects the chemical, physical and dynamic processes of rocks and their domains in the deep Earth, such as the relative movement of plates and the genesis and evolution of magmas [14–24]. Moreover, the amount of water in the continental lithospheric mantle is predicted to be closely related to its viscosity and stability [25,26]. Investigations on natural samples have demonstrated that an elevated water content can induce the destruction of cratons [27], whereas a reduced water content ensures craton longevity [22,28]. Therefore, understanding the concentration and distribution of water in the mantle plays a substantial role in explaining its formation and evolution. Furthermore, knowledge about the water content in the source of mantle-derived magmas may provide new constraints on the geochemical heterogeneities of the mantle and the involved processes [29,30]. The concentration and distribution of water in the deep lithosphere (lower crust and lithospheric mantle) and asthenospheric mantle can be determined by natural samples, including mafic granulites, peridotites, pyroxenites, eclogites and basalts, although information about the mantle transition zone (MTZ) and lower mantle is mostly constrained by high-temperature and -pressure experiments [31]. Recently, after a decade of research, a comprehensive dataset of the water content in peridotite and granulite xenoliths hosted by alkali basalts, terrain granulites, ultrahigh pressure metamorphic (UHPM) eclogites and Mesozoic-Cenozoic alkali basalts has been built for eastern China, which is now the only place with water-content data from such diverse types of natural samples (Fig. 1). This dataset provides an integrated picture of the water content and its distribution in the lower crust and upper mantle at a continental scale. In this review, we do not intend to compare our dataset from eastern China to those of other localities in the world because such comparisons have been covered well by Peslier [32] and Demouchy and Bolfan [33]. Instead, we will focus on the temporal and spatial distribution of water in the lower crust and upper mantle beneath eastern China and will summarize the origins of their heterogeneities and implications. Our primary purpose is to provide an example to address the water distribution and its role in the deep Earth at a continental scale. Figure 1. View largeDownload slide Sample locations in eastern China. (a) Peridotite and granulite locations; (b) basalt and eclogite locations. Figure 1. View largeDownload slide Sample locations in eastern China. (a) Peridotite and granulite locations; (b) basalt and eclogite locations. ANALYTICAL METHODS OF WATER CONTENT The main minerals in the Earth's upper mantle and lowermost crust include olivine (ol), clinopyroxene (cpx), orthopyroxene (opx), garnet (grt) and plagioclase (pl), and their high-pressure polymorphous. They are nominally anhydrous minerals (NAMs), meaning that there is no hydrogen in their stoichiometric chemical formulas. However, hydrogen can be incorporated into the structure of NAMs, usually as charge-compensating cation(s) often coupled with other cations (such as Al3+ or Ti4+). Hydrogen in such vacancies is normally bonded to one of the coordinating oxygens and forms a hydroxyl group [34], of which the content is expressed as H2O by weight ppm (hereafter referred to as ppm). Although the water content of minerals in the upper mantle and lowermost crust is generally less than several hundreds of ppm, they are likely the main water reservoir due to their predominant mass and volume proportions [32]. Fourier Transform Infrared Spectroscopy (FTIR) and Secondary Ion Mass Spectrometry (SIMS) are the most commonly used methods to measure the water content in NAMs [35]. Until now, all of the published water contents in mineral constituents of peridotites, granulites and eclogites from eastern China have been obtained using FTIR, with the exception of Aubaud et al. [36], who analysed several ol and pyroxene grains in Hannuoba peridotite xenoliths hosted in the Cenozoic basalts with SIMS. The SIMS data from Aubaud et al. [36] are comparable to the FTIR data of Yang et al. [37] from the same location. The water contents of the whole rocks, if reported, were estimated from the mineral water contents and their respective modal proportion through mass balance calculations. The water contents of basaltic magmas were calculated from the water content of cpx phenocrysts and the partition coefficient of water between cpx and basaltic melt. The details of the applied FTIR methodology can be found in Rossman [35], Kovács et al. [38], Xia et al. [27] and Demouchy and Bolfan [33]. The detailed run conditions and procedures have been described in the cited references. In short, the uncertainty of the measured water content of minerals in peridotites, granulites and eclogites was generally less than 20%, and that of the water content of basalts calculated from the water content of cpx phenocrysts was generally less than 40%. The factors influencing the uncertainty and the evaluation methods can be found in the cited references. WATER IN THE LITHOSPHERIC MANTLE OF EASTERN CHINA FTIR spectra and water contents The eastern Chinese continent consists of three main blocks from north to south: the Northeast China (NEC), the North China Craton (NCC) and the South China Block (SCB). Numerous small-volume basaltic volcanoes are distributed in these blocks and many contain abundant peridotite xenoliths, providing a good opportunity to investigate water distribution in the lithospheric mantle of eastern China. The water contents in the main constitute minerals (cpx, opx, ol and grt) in peridotite xenoliths from 28 basaltic localities, extending from Heilongjiang province in the north to Hainan province in the south (Fig. 1), have already been reported [37, 39–47]. Most of the studied peridotite xenoliths are spinel facies peridotites, except a few samples from the Nuomin volcano in the NEC and the Mingxi volcano in SCB, which are garnet facies peridotites [42,45]. FTIR investigations have shown that the garnets and most of the ol usually display no OH-related absorption bands, but there are also ol from several localities (Mingxi, Anyuan, Niutoushan and Qilin) in SCB that have prominent OH bands at 3575 cm−1, 3520 cm−1, 3340 cm−1 and 3230 cm−1. In contrast, except for the Nuomin peridotites, all of the opx and cpx show several OH absorption bands: 3595–3570 cm−1, 3525–3500 cm−1, 3415–3390 cm−1 and 3315–3300 cm−1 for opx, and 3635–3600 cm−1, 3550–3510 cm−1 and 3470–3445 cm−1 for cpx. The positions of these bands for ol, cpx and opx have been attributed to the vibration of the structural OH and are similar to those reported in peridotites worldwide [34,48–56]. It is particularly interesting that all of the ol, grt, cpx and opx grains in 13 peridotites from the Nuomin volcano in NEC have no any detectable OH band, demonstrating a very dry lithospheric mantle [45]. The water contents of minerals in the peridotites range from approximately 0 to 41 ppm for ol, 0 to 346 ppm for opx and 0 to 746 ppm for cpx. Except for some opx grains from the Panshishan and Tianchang basalts in the NCC [45,57], cpx and opx in peridotite xenoliths from other localities exhibit homogeneous water contents within individual grains, regardless of grain size. Moreover, the water content in opx usually displays positive correlations with the Al content [45,58]. Therefore, it is very likely that the cpx and opx retain the original water content that is typical for their mantle source. However, diffusional loss of hydrogen during peridotite ascent to the surface in the host basaltic magmas cannot be excluded for ol, considering near zero water in most of the ol grains. If Dcpx/ol = 10 is assumed, based on experimental results [20,21,36,39,59–63], the estimated water contents of the whole rocks using the calculated ol water content and the respective modal proportions of minerals range from 0 to 225 ppm, which cover the range of the Mid-ocean ridge basalt (MORB) source (50–200 ppm) [19,64,69] and are less than the ocean island basalt (OIB) source (300–1000 ppm) [68,70–74]. The water contents of the peridotites in eastern China vary among samples, even within a single locality. For example, the water contents range from 5 to 140 ppm, 5 to 355 ppm and 2 to 72 ppm in opx, cpx and whole rock, respectively, in peridotites from the Nushan volcano [37]. Furthermore, the water contents of the peridotites vary among different localities. For instance, the Nuomin peridotites in the NEC have no detectable water; the Hannuoba peridotites in the NCC have relatively low water content, ranging from 20 to 55 ppm, 50 to 150 ppm and 10 to 35 ppm for opx, cpx and whole rock, respectively; and the Jiande peridotites in the SCB contain much more water with 163 to 329 ppm, 388 to 589 ppm and 85 to 216 ppm for opx, cpx and whole rock, respectively [37,44]. Most of the ol grains in the peridotite xenoliths from eastern China have no detectable water; this is likely a result of the diffusional loss of hydrogen from ol during ascent, so the partition coefficient of water between ol and other minerals cannot be evaluated. The correlation of water content between cpx and opx is shown in Fig. 2 [36,40–47,58]. The ratios of H2O between cpx and opx fall in a range from 1.5 to 3.5 and can be regarded as partition coefficients (Dcpx/opx), since cpx and opx have preserved the original water content that is typical for their mantle source. The Dcpx/opx of the peridotites in eastern China has a similar range to those from natural peridotite xenoliths worldwide [22,34,50,52,53,55,56,75]. From the available experimental results [21,36,62,76,77], it appears that Dcpx/opx increases with increasing pressure (see the inserted plot in Fig. 2), so the variation in Dcpx/opx in peridotites worldwide may be a proxy for their depth of origin within the lithospheric mantle. If this is correct, it supports the common view of mantle petrologists and geochemists that the occasionally captured peridotite xenoliths by alkali magmas can well represent the entire lithospheric mantle, which is often questioned by geophysicists and geologists. Figure 2. View largeDownload slide Cpx water contents versus opx water contents in peridotite xenoliths of eastern China. The data are from Aubaud et al. [36], Yang et al. [58], Hao et al. [43–45,78], Li et al. [47], Wang et al. [46], Xia et al. [40,41] and Yu et al. [42]. The inserted plot shows the correlation between Dcpx/opx with pressure in experimental studies [21,36,62,76,77]. Figure 2. View largeDownload slide Cpx water contents versus opx water contents in peridotite xenoliths of eastern China. The data are from Aubaud et al. [36], Yang et al. [58], Hao et al. [43–45,78], Li et al. [47], Wang et al. [46], Xia et al. [40,41] and Yu et al. [42]. The inserted plot shows the correlation between Dcpx/opx with pressure in experimental studies [21,36,62,76,77]. Regional heterogeneity in the Cenozoic lithospheric mantle The regional distribution of the water content in the Cenozoic lithospheric mantle beneath all of eastern China, from the north part of the NEC (NNEC) to the SCB, has been discussed by Hao et al. [45] and is shown in Fig. 3. The Nuomin peridotite xenoliths are absolutely dry, indicating a dry lithospheric mantle beneath the NNEC. The lithospheric mantle of the south part of the NEC (SNEC) and the NCC is characterized by lower water contents than the MORB source, and the SNEC and the NCC share similar ranges and average values. The lithospheric mantle of the SCB has much higher water content than other regions in eastern China, falling in the range of the MORB source. By integrating the water content with other geochemical indices, Hao et al. [45] suggested that the regional variations in the water content of the lithospheric mantle beneath eastern China cannot be caused by distinct partial melting and mantle metasomatism events or the redox state. Instead, the lithospheric mantle beneath different regions may have distinct origins and have undergone distinct geodynamic processes. The NNEC lithospheric mantle is supposed to be from the Siberia craton, with hydrogen diffused out of the peridotite minerals when the Siberia craton had interacted with a super mantle plume. The relatively low water content of the NCC and the SNEC may have been caused by the reheating effect of an upwelling asthenospheric flow during the lithospheric thinning event in the Mesozoic. The higher water content of the SCB peridotites suggests that the large part of its deeper lithospheric mantle was accreted from the asthenosphere, and either the SCB did not undergo a significant lithospheric thinning event such as the NCC or the thinning mechanism was different. From the case of eastern China, it seems that water distribution in the lithospheric mantle may provide new constraints on the origins and geodynamic processes of continents. Figure 3. View largeDownload slide Comparison of the water contents of cpx, opx and whole rock of the peridotite xenoliths in different blocks from eastern China. The data sources are the same as in Fig. 2. Figure 3. View largeDownload slide Comparison of the water contents of cpx, opx and whole rock of the peridotite xenoliths in different blocks from eastern China. The data sources are the same as in Fig. 2. Temporal heterogeneity in the NCC lithospheric mantle Although the spatial distribution of water content and the controlling factors in the lithospheric mantle are of increasing interest to the Earth science community, less attention has been paid to the temporal variation of water content, which also matters greatly and may provide new insights into the evolution of the lithospheric mantle. Abundant mantle xenoliths entrained in the alkali basalts with a wide range of eruption ages (125 to < 40 Ma) across the NCC, especially the eastern part, provide a good chance to address that issue. A detailed investigation of the bulk water content of peridotite xenoliths from many localities indicates that the Cenozoic lithospheric mantle (<40 Ma) beneath the eastern part of the NCC was relatively dry, with an average of 25 ± 18 ppm (from 15 to 85 ppm), except for a few localities close to the deep-cutting Tan-Lu fault [40,45,58]. The extremely low bulk water content below 50 ppm compared to the samples from other continental lithospheric mantles worldwide that are typical for cratonic and off-cratonic peridotites (normally 40–180 ppm, with average values of 119 ± 54 ppm and 78 ± 45 ppm, respectively) and sub-oceanic mantle lithosphere values (>50 ppm) inferred from MORB and OIB (see the compiled dataset in Xia et al. [40]) highlights the uniqueness of the eastern part of the NCC and implies its links with the destruction of the NCC, as stated in the previous section. Li et al. [47] reported the water content of the peridotite xenoliths hosted by the late Mesozoic basalts (82–67 Ma) from eastern NCC, revealing two types of mantle domains with contrasting bulk water contents. The lithospheric mantle domain, represented by the Daxizhuang xenoliths at 74 Ma, was relatively ‘dry’, sharing similar characteristic with the Cenozoic lithospheric mantle [40,47,58]. Meanwhile, the peridotite xenoliths from Junan (67 Ma) and Qingdao (82 Ma) suggest a relatively high bulk water content for the lithospheric mantle, with an average of 130 ± 20 ppm, which is in the range of the MORB source (50–250 ppm). When mantle xenoliths are scarce, the bulk water content of the lithospheric mantle can also be obtained indirectly. Based on a study of water in the early crystallized cpx phenocrysts in basalts, Xia et al. [27] estimated that the bulk water content of the lithospheric mantle source of the Feixian high-magnesium basalts. The bulk water content was more than 1000 ppm, which is several times higher than that of the MORB source, demonstrating that the NCC lithospheric mantle was very water-enriched in the early Cretaceous (∼120 Ma), at the paroxysm of the destruction of the NCC. These combined snapshots envisage a temporal variation of water content for the eastern NCC lithospheric mantle (Fig. 4) [47]. At the peak of the destruction (∼120 Ma), the lithospheric mantle was hydrous; after its destruction was complete at the end of the Mesozoic, the water content of the lithospheric mantle was extremely low; during the period of destruction, the lithospheric mantle had intermediate water content [47]. Li et al. [47] further discussed the process of this cratonic destruction as follows: (i) at the peak time of destruction, the hydrated, water-rich lithospheric mantle offers a presumably low-viscosity state for its own destruction; (ii) during the destruction, the upwelling asthenosphere erodes the lowermost lithospheric mantle and continuously heats the overlying lithospheric mantle, which results in lithospheric thinning and dehydration of the relict lithospheric mantle. This goes on until the cessation of the destruction of the cratonic lithospheric mantle, which is, at least in part, due to its now lower water content and stiffer rheology, therefore supporting better resistance to further tectonic removal. The upwelling asthenosphere, however, cools and is transformed into a newly accreted lithospheric mantle with probably higher bulk water content than that of the dehydrated relict cratonic lithospheric mantle. Figure 4. View largeDownload slide The temporal variation of the bulk water content in the lithospheric mantle beneath the eastern NCC. The figure is modified from Fig. 11 in Li et al. [47] by adding the cpx water contents and the data sources are the same as Li et al. [47]. Figure 4. View largeDownload slide The temporal variation of the bulk water content in the lithospheric mantle beneath the eastern NCC. The figure is modified from Fig. 11 in Li et al. [47] by adding the cpx water contents and the data sources are the same as Li et al. [47]. These findings above seem to contrast with those of Liu and Xia [79], who reported the water content of six peridotite xenoliths entrained in the early Cretaceous (∼125 Ma) high-Mg diorite from Fushan in the western part of the NCC, which show much lower water content than their hydrated, more water-rich Mesozoic counterparts in the eastern part of the NCC. The water content in cpx and opx ranges from 216 to 404 ppm and 123 to 188 ppm, respectively, with an average of ∼40 ppm for whole rock, which is far less than >1000 ppm of the Feixian mantle source but is in the range of the classic craton [34,50,80,81]. The contrasting water contents could be interpreted as the hydration of the NCC lithospheric mantle by the westward subduction of the paleo-Pacific plate, which may not have yet reached the western part of the NCC by the early Cretaceous. WATER IN THE LOWER CONTINENTAL CRUST OF EASTERN CHINA Temporal variations in water content of the lower continental crust The lower continental crust, separating the shallow crust from the underlying lithospheric mantle, is of crucial importance in determining the tectonic evolution of continents and in buffering the exchange of materials between Earth's interior and exterior. Despite great efforts of water in mantle minerals, much less attention has been paid to the speciation and concentration levels of water in the lower crust and its bearing on physical properties. A survey by FTIR on xenolith and terrain granulites, typical of samples from the lower crust, from Hannuoba, Nushan and Daoxian, in eastern China, has shown that the main constitutive minerals, such as plagioclase (pl), cpx, opx and grt, commonly contain trace amounts of OH in their lattice structure, also including molecular H2O for some pl [58,82]. Thus, minerals in the lower crust have a similar ability to dissolve water as minerals in the upper mantle, although its concentration may be quite different (as shown below). The calculated water concentrations of lower crustal granulites range from approximately 200 to 2000 ppm for cpx, 60 to 1800 ppm for opx, 65 to 900 ppm for pl, 300 to 600 ppm for grt and 150 to 1100 ppm for the bulk rocks (Fig. 5) [37]. Recently, Németh et al. [83] measured the water contents of minerals in a suite of granulite xenoliths from the Pannonian Basin in east-central Europe; their data demonstrate the occurrence of approximately 0–271 ppm, 55–413 ppm and 0–684 ppm for cpx, opx and pl, respectively. These values are much smaller than those of other granulites from eastern China. Therefore, it appears that the distribution of water is highly heterogeneous in the lower crust, on both small (i.e. different samples from the same locality) and large (i.e. different localities) scales. This is similar to the observed chemical heterogeneities of many elements and isotopes in samples from the lower crust (e.g. Rudnick and Gao [84] and references therein). Such heterogeneities are presumably either inherited from the protoliths or they reflect the complicated history of crust–mantle interactions. Figure 5. View largeDownload slide Water contents of the lower crustal minerals in eastern China and their temporal contrast (after Yang et al. [37]). Figure 5. View largeDownload slide Water contents of the lower crustal minerals in eastern China and their temporal contrast (after Yang et al. [37]). An interesting result arising from the available data is that, by classifying the samples according to their formation ages, pre-Phanerozoic granulites record apparently higher water contents than Phanerozoic granulites (Fig. 5) [37]. One possible cause is that the early lower crust was more hydrous than the modern one, although it is not clear whether this is a local phenomenon or a general trend on a global scale [37]. For the granulite samples compiled in Fig. 5, the chemical composition is broadly similar for individual minerals, and so are the temperature and pressure conditions of equilibrium [37]. Therefore, a significant contrast in water content in these samples deserves some further discussion, even if Yang et al. [37] already provided some clues from a petrological perspective. Recently, Yang [85] experimentally demonstrated how, at a given pressure and temperature, the ability of pl in incorporating OH is more significant under very reduced conditions, e.g. at four or more log units below the QFM (quartz-fayalite-magnetite) buffer. Considering the thermodynamics of OH dissolution in silicate minerals [86] and the redox state of around QFM in the modern lower crust [87], the observed water contrast in Fig. 5 can be explained by a change of the prevailing redox state in the lower crust, in that the pre-Phanerozoic lower crust of eastern China was regionally more reduced. Alternatively, it is possible, considering the even lower water contents of granulite xenoliths from the younger Miocene Pannonian Basin in Central Europe [83], that the observed lower water content in the Phanerozoic Chinese xenolith may be related to repeated depletion events since the Precambrian. This appears to be a logical explanation, as the continental crust may have remained stable over geologic time, whereas the continental lithospheric mantle may have been removed during cratonic destruction. This means that the water budget in the continental lower crust may have decreased gradually over geologic time during repeated melt removals and heating episodes during extension and that it could only be rehydrated locally, as suggested by Németh et al. [83]. Vertical distribution of water content in the lower crust and lithospheric mantle The vertical distribution of water content in the shallow mantle has received increasing attention in recent years, and has been addressed in some studies [28]. However, the variation in water content from the lithospheric mantle to the overlying lower crust has not been well documented, even if it may be of equal importance for the stability and tectonics of the continents. Yang et al. [58] addressed this question with reported data of water contents from both granulite xenoliths and peridotite xenoliths from Hannuoba and Nushan within the NCC (eastern China). They found much higher water contents in both the mineral constituents and bulk rocks in granulites than in peridotites (Fig. 6) [58]. The contrast of water content may be related to petrological processes and histories of these rocks, such as partial melting, differentiation and metamorphism; however, thermodynamically, it may indicate that the incorporation mechanism of OH in lower crustal minerals is different from the mechanism in mantle minerals, e.g. due to different pressure, temperature and mineral chemistry (lower crustal pyroxenes are richer in Fe and Al than their mantle counterparts) [58]. Figure 6. View largeDownload slide Vertical distribution of water content in the lower crustal granulite xenoliths and mantle peridotite xenoliths. (a) The cpx and opx in the Hannuoba xenoliths; (b) the cpx and opx in the Nushan xenoliths; and (c) the bulk water content (after Yang et al. [58]). Figure 6. View largeDownload slide Vertical distribution of water content in the lower crustal granulite xenoliths and mantle peridotite xenoliths. (a) The cpx and opx in the Hannuoba xenoliths; (b) the cpx and opx in the Nushan xenoliths; and (c) the bulk water content (after Yang et al. [58]). The vertical variation in water content in the lower crust and sub-continental lithospheric mantle and their lateral differences, as shown in Fig. 6, allows us to explore the rheological properties of the deep lithosphere in eastern China. Considering the data on water and regional thermal (P–T) conditions, it appears that the lower crust is much stronger than the underlying shallow lithospheric mantle at Hannuoba but weaker at Nushan. This makes the lithosphere thermally and mechanically unstable at Hannuoba compared to Nushan. This indicates that, during the Cenozoic when the xenoliths were brought to the surface, the lithosphere may have undergone mechanical thinning at Hannuoba but thickening at Nushan [58]. This means that different lithospheric processes may have been operative in different tectonic areas beneath eastern China and that the variation in water content in the deep lithosphere is critical for understanding deep processes. WATER IN THE CENOZOIC ALKALI BASALTS OF EASTERN CHINA In eastern China, from Wudalianchi in Heilongjiang Province in north-eastern China to the Hainan Island in the south end of mainland China, the Cenozoic basalts are widely distributed. These basalts generally exhibit typical OIB-like trace element patterns and Sr–Nd isotopic compositions. The enriched components in the mantle sources of the Cenozoic basalts in eastern China are still not fully understood. The old lithospheric mantle, the recycled ancient lower continental crust and recycled oceanic crust in the asthenosphere were all suggested to be responsible for their formation [88–93]. Recently, the materials in the hydrated MTZ beneath eastern Asia were also proposed to be involved in the mantle sources of the basalts in north-eastern China [94,95]. It has been widely accepted that the water content of magma, and the corresponding H2O/Ce ratios, would be useful for identifying the mantle components [96–98]. The importance of water in our understanding of the genesis of alkali basalts in eastern China was also invoked as early as 2007 [99]. However, there has been minimal research trying to analyse the water contents of these basalts, except several attempts to argue that basalts from several locations may be rich in water that used thermodynamic calculations [95,100] or an indirect inference from the low δ18O values of mineral phenocrysts [101]. Only very recently were the water contents of the Cenozoic alkali basalts, from north-eastern to south-eastern China, constrained [29,30,102–105]. These results are based on a method that relies on the water contents of cpx phenocrysts and chemical composition-dependent water partitioning between cpx and basaltic melts, which was first suggested by Wade et al. [106] and considerably improved by Xia et al. [27] and Liu et al. [30]. Here, we summarize the distribution of the water contents in these alkali basalts in eastern China determined by this method and its inferences about the sources of the basalts, enriched components in the sources and implications about the role of the subducted Pacific slab on the genesis of the Cenozoic basalts in eastern Asia. Regional distribution of H2O and H2O/Ce It is worth mentioning that the uncertainty of this methodology may be up to ∼40% [30,41]. To recover the water content of the Cenozoic alkali basalts, cpx phenocrysts with Mg# (=100* Mg/(Mg+Fe), mol percent) >75 were used [29,30,104]. Although multiple processes, such as crystallization, crustal contamination, degassing and diffusion of hydrogen out of cpx during or after eruption, might affect the final calculated magma-water contents, reasonable correlations between the calculated water contents and incompatible element contents of bulk rocks (Fig. 7 [29,103] shows examples from the Shandong and Shuangliao basalts) indicate that none of these processes may have a significant impact [29,104]. The H2O/Ce ratios were determined by the calculated water contents and the measured Ce concentration of the bulk rocks, which are rather consistent with the average H2O/Ce values calculated by $$\frac{{C_w^{cpx}/D_w^{cpx/melt}}}{{C_{Ce}^{cpx}/D_{Ce}^{cpx/melt}}}$$ for a group of cpx phenocrysts, where C is the concentration of H2O and Ce, and D is the partition coefficient between cpx and basaltic melt [104]. Figure 7. View largeDownload slide The correlation between the calculated magma water contents and the bulk rock trace element concentrations. The data for the Shandong and Shuangliao basalts are from Liu et al. [104] and Chen et al. [29], respectively. Figure 7. View largeDownload slide The correlation between the calculated magma water contents and the bulk rock trace element concentrations. The data for the Shandong and Shuangliao basalts are from Liu et al. [104] and Chen et al. [29], respectively. The water contents and H2O/Ce ratios of the alkali basalts from different locations in eastern China are statistically shown in Fig. 8 [29,30,64–68,70,71,74,96–98,102–104,107–119], from which local and regional scale heterogeneities can be identified. The recovered magma-water contents varied from 0.6 to 3.9 wt.% in the Shandong basalts, from 0.9 to 3.1 wt.% in the Shuangliao basalts, from 1.1 to 2.7 wt.% in the Zhejiang basalts, from 0.2 to 3.8 wt.% in the Fujian basalts and from 1.6 to 4.2 wt.% in the Guangdong basalts. The higher water contents fall in the range of back-arc basin basalts (BABB) and island arc basalts (IAB), even considering the ∼40% uncertainty. By contrast, the water contents of the basalts in Taihang and Wulanhada in Central NCC and Xiaogulihe in the northern part of NE China are much lower (0.2–1.4 wt.%, 0.21–0.69 wt.% and ∼0.5 wt.%, respectively), which are close to the water contents of MORB and typical OIB. Note, the water contents found in the Wulanhada basalts do not support the conclusion that the basalts in this area are very hydrous [100,120], which was deduced from non-analytical approaches (the MELTs program modeling or indirect evidence from the involvement of hydrated lower oceanic crust, respectively). When the results from direct measurements are considered, it seems that the basalts in the region closer to the Pacific trench (Shuangliao, Shandong, Zhejiang, Fujian and Guangdong, referred to as the eastern area in Fig. 9) [120] exhibit higher water content in the magma than those further away from the trench (Xiaogulihe, Wulanhada and Taihang, referred to as the western area in Fig. 9). However, as an outlier, the water contents of the Chaihe-Aershan basalts in NE China span a similar range with those of the Shandong and Shuangliao basalts, although they have comparable distance to the Pacific trench as to the Wulanhada and Xiaogulihe basalts. Correspondingly, the H2O/Ce ratios of basalts exhibit a similar regional distribution (Fig. 8). When the maximum and mean values of the H2O/Ce ratios of the Shandong, Shuangliao, Zhejiang, Fujian, Guangdong and Chaihe-Aershan basalts are obviously higher than those of the most water-rich OIB (∼250) [98], those of the Taihang, Wulanhada and Xiaogulihe basalts are close to or significantly lower than that of the normal MORB (∼200–210) [122]. Figure 8. View largeDownload slide The water contents and H2O/Ce ratios of the Cenozoic alkali basalts in eastern China (after Chen et al. [121]). The data are presented from northern to southern localities from top to bottom. The water content data of the Cenozoic basalts in eastern China are from Chen et al. [29,102,103,121], Liu et al. [30,104] and Liu et al. [105]. The data source for IAB, BABB, OIB, MORB and E-MORB is from Asimow et al. [64], Danyushevsky et al. [107,108], Dixon and Clague [97], Dixon et al. [65,70,98], Dobson et al. [109], Hochstaedter et al. [110], Michael [66,96], Nichols et al. [71], Saal et al. [67], Simons et al. [68], Sisson and Layne [110], Sobolev and Chaussidon [69], Stolper and Newman [112] and Wallace [113,114]. The H2O/Ce ratio of PM is calculated with H2O from Dixon and Clague [97] and Ce from Sun and McDonough [115], and the H2O/Ce ratio of the EM1, EM2, HIMU, FOZO, DMM and IAB is from Cabral et al. [116], Dixon et al. [98], Kendrick et al. [117,118], Workman et al. [74] and Ruscitto et al. [119]. Figure 8. View largeDownload slide The water contents and H2O/Ce ratios of the Cenozoic alkali basalts in eastern China (after Chen et al. [121]). The data are presented from northern to southern localities from top to bottom. The water content data of the Cenozoic basalts in eastern China are from Chen et al. [29,102,103,121], Liu et al. [30,104] and Liu et al. [105]. The data source for IAB, BABB, OIB, MORB and E-MORB is from Asimow et al. [64], Danyushevsky et al. [107,108], Dixon and Clague [97], Dixon et al. [65,70,98], Dobson et al. [109], Hochstaedter et al. [110], Michael [66,96], Nichols et al. [71], Saal et al. [67], Simons et al. [68], Sisson and Layne [110], Sobolev and Chaussidon [69], Stolper and Newman [112] and Wallace [113,114]. The H2O/Ce ratio of PM is calculated with H2O from Dixon and Clague [97] and Ce from Sun and McDonough [115], and the H2O/Ce ratio of the EM1, EM2, HIMU, FOZO, DMM and IAB is from Cabral et al. [116], Dixon et al. [98], Kendrick et al. [117,118], Workman et al. [74] and Ruscitto et al. [119]. Figure 9. View largeDownload slide (a) Comparison of the water contents in the sources of the Cenozoic eastern China basalts with those in the peridotite xenoliths from the same regions. (b) The spatial distribution of the water content in the sources of the Cenozoic eastern China basalts. After Chen et al. [121]. Figure 9. View largeDownload slide (a) Comparison of the water contents in the sources of the Cenozoic eastern China basalts with those in the peridotite xenoliths from the same regions. (b) The spatial distribution of the water content in the sources of the Cenozoic eastern China basalts. After Chen et al. [121]. Where did water come from? Whether the Cenozoic basalts in eastern China are from the lithospheric mantle is still under fierce debate after more than three decades of intensive study [30,89,91,92,104,123–127]. Systematic investigations of the water contents of the alkali basalts and their hosted peridotite xenoliths that represent the lithospheric mantle may provide new insights. More precisely, our dataset on the water contents of eastern Chinese samples allow us to make both local and regional comparisons. This means that the water-content contrast between the basalt source and the lithospheric mantle for a specific area, and the regional water distribution in the lithospheric mantle and in the basalt source for entire eastern China could be explored. Chen et al. [121] have estimated the water contents in the sources of the Cenozoic eastern China basalts using the following: (i) the calculated water contents of the initial basaltic melts; (ii) the available partition coefficients of water between basalts and mantle rocks; and (iii) the estimated partial melting degrees that the basalt source experienced. The mantle source-water contents ranged between approximately 150 and 4700 ppm and most of them were higher than 500 ppm (Fig. 9b), regardless of the used partial melting model (batch melting or fractional melting) [121]. By comparing the estimated water contents in the basalt sources with those in the peridotite xenoliths [121], the flowing can be clearly determined: (i) the source-water contents for all the basalts are significantly higher (several to several hundred times) than those of the lithospheric mantle in the same region (Fig. 9a); (ii) except for the Chaihe-Arershan basalts in the north-west NEC, the water contents of the basalt sources appear to decrease from the east to the west (Fig. 9b), which is different from the water-content trend in the lithospheric mantle of eastern China decreasing from the south to the north (Fig. 3). Thus, the local and regional differences in the water contents in the lithospheric mantle and the basalt sources indicate that the Cenozoic basalts in eastern China may not have been derived from the lithospheric mantle. The upper mantle (except the lithospheric mantle containing amphibole or mica) can only accommodate 50–250 ppm of H2O [19,31,128], and the lower mantle contains less or similar amounts of water [128,129]. Only the MTZ can contain several hundred ppm to >10 000 ppm H2O [128,130]. Therefore, the estimated water contents (150–4700 ppm, and most of them higher than 500 ppm) in the sources of the Cenozoic eastern China basalts are likely from the MTZ. This proposition is compatible with geophysical investigations that have shown the stagnated oceanic slabs in the MTZ beneath eastern Asia [131,132], which are expected to provide water once they are disturbed. Enriched components in the mantle source The magma-water contents and corresponding H2O/Ce ratios provided new constraints on the origin of enriched components in the mantle sources of the Cenozoic eastern China basalts. As to the significance of the magma-water contents for identifying the source components, one typical example is the case of the Cenozoic alkali basalts in Shandong [104]. Both the recycled Pacific oceanic sediments and residual ancient lower continental crust (in the form of eclogite) that was produced by an earlier partial melting event have been suggested to be enriched components in the mantle source of the Cenozoic Shandong basalts [89,91,92,125]. Liu et al. [104] reported the water contents of the alkali basalts in one volcano (Fuyanshan volcano) in the same region. In addition to their exceptionally high water contents (see Regional distribution of H2O and H2O/Ce), the H2O/Ce ratios of these basalts were positively correlated with bulk-rock 87Sr/86Sr ratios and were negatively correlated with Nb/U ratios, respectively (Fig. 10) [104]. These trends demonstrate that the enriched components in the basalt sources should be enriched with water. Thus, the dry lower continental crust that experienced an earlier melting event before incorporation into the mantle sources could not be the appropriate candidate. Figure 10. View largeDownload slide The correlation of H2O/Ce with the bulk-rock 87Sr/86Sr and Nb/U ratios of the Fuyanshan basalts in Shandong. Modified from Liu et al. [104]. Figure 10. View largeDownload slide The correlation of H2O/Ce with the bulk-rock 87Sr/86Sr and Nb/U ratios of the Fuyanshan basalts in Shandong. Modified from Liu et al. [104]. The role of the Pacific subduction in the genesis of Cenozoic basalts in eastern China Many trace elemental, radiogenic and stable isotopic evidences have been found to support the mantle sources of basalts from ∼110 Ma to the late Cenozoic in eastern China that contain components from subducted oceanic slabs [91,101,133,134]. This hypothetical oceanic slab has usually been assigned to the subducted Pacific slab, which was stagnant at the MTZ beneath eastern Asia [131,135]. Recently, the changes in basalt geochemistry (such as Eu/Eu*, 87Sr/86Sr and δ18O values) with eruption ages were determined [92,126,136–138] and they were suggested to reflect the secular contributions of different portions of the subducted Pacific oceanic crust in the mantle sources. Very recently, similar temporal variations in magma-water contents for several Cenozoic basalts in eastern China were resolved. Chen et al. [29] reported the water contents of the Shuangliao basalts in NNEC, which erupted from 50 to 43 Ma. As shown in Fig. 11, the younger basalts exhibit higher H2O/Ce ratios along with higher Ba/Th and lower Ce/Pb ratios, which reflect a secular contribution of the subducted oceanic sediments in the mantle sources. The similar temporal variations of H2O/Ce and other trace elemental ratios were also observed for the basalts in the Chaihe-Aershan (∼1.27–0.25 Ma) volcanoes of north-east China and the Zhejiang (26–17 and 11 Ma) volcanoes of SCB [103,105]. Overall, all those recycled oceanic components in the mantle sources of the Cenozoic basalts were dynamically incorporated, which would be best linked to the on-going subduction of the Pacific slab. Figure 11. View largeDownload slide Correlations among H2O/Ce of the Cenozoic Shuangliao basalts and the bulk rock trace elemental ratios and radiogenic isotopic compositions. Gray squares represent the possible end members in the mantle source. Modified from Chen et al. [29]. Figure 11. View largeDownload slide Correlations among H2O/Ce of the Cenozoic Shuangliao basalts and the bulk rock trace elemental ratios and radiogenic isotopic compositions. Gray squares represent the possible end members in the mantle source. Modified from Chen et al. [29]. WATER IN THE DABIE-SULU UHPM ECLOGITES: MEANS OF CARRYING CRUSTAL WATER INTO THE DEEP Water content and distribution in the Dabie-Sulu UHPM eclogites It is well known that a great deal of aqueous fluid can be released from altered basalts and overlying sediments of the oceanic crust with increasing temperature and pressure during plate subduction. This process hydrates the overlying mantle wedge and usually generates syn-subduction arc magmatism on a large scale [139,140]. In contrast, aqueous fluids are less significant or absent during ultrahigh pressure metamorphism (UHPM) of the continental crust at mantle depths relative to that of the oceanic crust [141]. Because fluid is an important agent for element and isotope exchange, its presence or absence plays a critical role in the attainment and preservation of geochemical equilibrium between metamorphic minerals, and even the re-equilibration of geochronometric and geothermobarometric systems [142]. The Dabie-Sulu orogenic belt in east-central China is the biggest exposed UHPM belt in the world, providing a natural laboratory to investigate the transportation and recycling of water. In principle, the major constituent minerals of UHPM eclogites are NAMs such as garnet, omphacite, rutile and silica phases. They are all stable phases in the P–T range typical of continental subduction-zone metamorphism and likely represent the primary water reservoirs in subducted slabs. FTIR analyses of NAMs in the Dabie-Sulu UHPM eclogites demonstrate that they contain a certain amount of water as structural hydroxyl locked in vacancies, which can be up to ∼1915 ppm for garnet, ∼1850 ppm for omphacite, ∼9590 ppm for rutile and ∼440 ppm for coesite [143–148]. The similar quantities of water in both garnet and omphacite indicate that they play equivalent roles in transporting surface water to mantle depths during continental subduction (Fig. 12a) [34,143–157]. Similarly to oceanic subduction, such transported water by subducted continents would cause prominent hydration (at least locally) in the mantle which they interact with. Figure 12. View largeDownload slide Water contents in the Dabie-Sulu eclogites and comparisons with other UHPM eclogites and eclogite xenoliths. (a) water contents in garnet, omphacite and bulk rocks from the Dabie-Sulu orogenic belt; (b) water contents in garnets from the Dabie-Sulu eclogites and other UHPM eclogites and eclogite xenoliths; (c) water contents in omphacites from the Dabie-Sulu eclogites and other UHPM eclogites and eclogite xenoliths. The dataset for the Dabie-Sulu eclogites is from Zhang et al. [143,144], Xia et al. [145], Sheng et al. [146], Chen et al. [147], Zhao et al., [148] and Zhao and Zhang [149]; and the dataset for other UHPM eclogites and eclogite xenoliths is from Aines and Rossman [150], Bell and Rossman [34], Langer et al. [151], Snyder et al. [152], Matsyuk et al. [153], Ragozin et al. [154], Bell et al. [155], Katayama et al. [156] and Skogby et al. [157]. Figure 12. View largeDownload slide Water contents in the Dabie-Sulu eclogites and comparisons with other UHPM eclogites and eclogite xenoliths. (a) water contents in garnet, omphacite and bulk rocks from the Dabie-Sulu orogenic belt; (b) water contents in garnets from the Dabie-Sulu eclogites and other UHPM eclogites and eclogite xenoliths; (c) water contents in omphacites from the Dabie-Sulu eclogites and other UHPM eclogites and eclogite xenoliths. The dataset for the Dabie-Sulu eclogites is from Zhang et al. [143,144], Xia et al. [145], Sheng et al. [146], Chen et al. [147], Zhao et al., [148] and Zhao and Zhang [149]; and the dataset for other UHPM eclogites and eclogite xenoliths is from Aines and Rossman [150], Bell and Rossman [34], Langer et al. [151], Snyder et al. [152], Matsyuk et al. [153], Ragozin et al. [154], Bell et al. [155], Katayama et al. [156] and Skogby et al. [157]. Heterogeneous water contents in eclogite minerals were observed not only on the scale of the outcrop, but also among the different grains within the same hand-specimen. For instance, a >1000-ppm variation occurred for four to five garnet grains from ∼1-cm eclogite section [145]. Such small-scale heterogeneity clearly suggests very limited mobility of fluids during UHP metamorphism, and both subduction and exhumation processes of UHP rocks occurred over a short period of time [145]. Moreover, the decreased water content in the rims of some garnet and omphacite grains compared with the cores have been ascribed to hydroxyl exsolution upon the initial decompression exhumation of the UHPM eclogites. This process provides an efficient way to generate fluids during the early stage of exhumation and also promotes retrograde metamorphism [146]. The bulk-rock water contents are estimated to lie between 100 and 750 ppm [143–146]. In addition to structural OH, mineralogical studies by means of FTIR, transmission electron microscopy (TEM), backscattered electron (BSE) and laser Raman techniques indicate that NAMs contain significant amounts of water in the form of molecular H2O in fluid inclusions [145,158–161]. The fluid inclusions hosted by UHPM minerals may be either primary or secondary [162], thus providing an upper limit for the ability of transporting water into the mantle depth, while the structural hydroxyl in NAMs defines the lower limits. Comparison with upper-mantle eclogites It is worth noting that either consistency or inconsistency of water contents has been observed in UHPM eclogites relative to mantle eclogite xenoliths hosted by kimberlites. For garnets, the water contents of UHP metamorphic rocks are significantly higher than in eclogite xenoliths (Fig. 12b). The prolonged residence in mantle conditions of eclogite xenoliths has the potential to result in a significant loss of water to the surrounding peridotitic mantle due to the usually much lower water contents in upper-mantle peridotites. Consequently, this can explain the lower water contents in garnets from eclogitic xenoliths [145]. In contrast, due to the presumably much shorter duration of tectonic exhumation, the UHPM eclogites may have a better capacity to preserve the water-rich character of their protoliths [145,146,163]. However, the water contents in omphacites from UHPM eclogites are comparable to those in eclogite xenoliths (Fig. 12c). This may reflect both the pressure and composition control on the incorporation of structural hydroxyl in crystal defects such as the Ca-Eskola components in M2-site vacancies [157,164]. Moreover, jadeite with water contents ranging from 100 to 1950 ppm and coherent variations in Na and Ca contents, as well as M2-site vacancies, have been reported for the Dabie UHPM jadeite-bearing quartzites [165]. These characteristics suggest that bulk mineral composition plays an important role in the incorporation of hydrogen [165]. Taken as a whole, the UHPM eclogites can contain higher water contents (>200 ppm) than MORB-sourced mantle peridotites that range from 50 to 200 ppm [166]. Therefore, the breakdown of such a water-rich subducted continental slab would have the potential to contribute to the ‘hidden’ water-rich region in mantle depths as the OIB source. Water in the form of structural hydroxyl is usually immobile in both hydrous and anhydrous minerals in UHPM conditions. The breakdown of the hydrous minerals in UHPM slices provides a dominant source for aqueous fluid during continental subduction-zone metamorphism [167,168]. Furthermore, numerous experimental studies have demonstrated that the solubility of structural hydroxyl in NAMs increases with pressure [156,164,169–175]. This, in turn, implies that the structural hydroxyl can be released from NAMs during exhumation. Therefore, significant amounts of aqueous fluids are available from the exsolution of structural hydroxyl and molecular water from NAMs in the initial stage of exhumation. The aqueous fluid then reacts with its host minerals resulting in their dissolution and recrystallization. Consequently, this metamorphic dehydration during subduction/exhumation would probably provide a sufficient amount of aqueous fluid for not only high-pressure eclogite-facies quartz veining, but also amphibolite-facies retrogression. SUMMARIZED CONCLUSIONS Although the main minerals in the lithospheric mantle of eastern China are nominally anhydrous, they usually contain a certain amount of water (up to several hundreds of ppm H2O) as hydroxyl groups in their structural defects. The temporal variation of the water content in the lithospheric mantle from the early Cretaceous (∼120 Ma) to Cenozoic (<40 Ma) appears tightly connected to the stability of the NCC (from its destruction to its consolidation), thus providing a clear case to show that the rheological changes induced by water in the lithospheric mantle affect the stability of continents. The heterogeneous water content in the Cenozoic lithospheric mantle beneath different blocks of eastern China was not only inherited from tectonic settings from which they came before having been amalgamated into a single continent, but was also affected by later geological processes that they experienced. The distinct water content between the lowermost crust and lithospheric mantle of eastern China and its induced rheological contrast at the base of the crust indicate that the continental crust–mantle boundary could behave either in a coupled or decoupled style beneath different areas and/or at different stages. The water contrast between the Precambrian and Phanerozoic lower crust of eastern China suggests a temporal evolution of the water content in the Earth's crust, different formation mechanisms of the continental lower crust or both. It is likely that the continental lower crust may have become gradually depleted with respect to its water content over geologic time, contributing to the stronger rheology and, therefore, preservation of the continental crust. The alkali basalts of eastern China demonstrate a heterogeneous distribution of the water content in the upper mantle and MTZ. Pacific plate subduction is likely the main process to introduce heterogeneities in the water content and other geochemical characteristics of the intra-plate magmas. Local and regional comparisons of the water content between the lithospheric mantle and basalt source demonstrate that the Cenozoic alkali basalts in eastern China were not sourced from the lithospheric mantle. Instead, the inferred high water contents in the mantle sources indicate that the Cenozoic eastern China basalts were likely sourced from the MTZ. The UHPM eclogites of the Dabie-Sulu orogeny can host up to thousands of ppm of water in nominally anhydrous garnets and omphacites, demonstrating that the ‘dry’ continental crust may carry a certain amount of water back into the deep mantle by subduction. Such recycled continental crustal materials not only created a local water-rich zone, but also delivered continental geochemical signatures into the mantle, both accounting for the continental geochemical imprints in some OIBs and other intra-plate basalts. FUTURE ISSUES More data from other areas are necessary to examine certain conclusions drawn from the samples of eastern China, such as the higher water content in the lower crust than in the underlying lithospheric mantle and the more hydrous Precambrian lower crust than its Phanerozoic counterpart. In addition, the spatial and temporal distribution of the water content in the lithospheric mantle and its effect on continental evolution also require further exploration by collecting more data from other parts of the world. Pyroxenites are an important rock constituent in the upper mantle, and are likely the source lithology of some intra-plate basalts; in particular, local enrichment of Fe- and H-bearing pyroxenites may lead to regionally electrical anomalies, affecting the structure and some key properties of the upper mantle [176]. However, less attention has been paid to water in pyroxenites, and so far only Bizmis and Peslier [177] have reported water in six pyroxenite xenoliths with similar genesis (crystallized from basaltic magmas) from Hawaiian basalts. Due to the prominent role of water in the temperature, pressure and degree of melting and composition of the yielded melts, it is necessary to systematically study water in pyroxenites in different tectonic settings. Felsic rocks are the main constituents in UHPM orogens and the dominant assemblages in the shallower crust and the content and partitioning of water in minerals of felsic rocks should be investigated for a more comprehensive understanding of the recycling of water during the subduction and exhumation of continental plates and the storage of water in the crust and its exchange between different reservoirs. The magma-water contents provide new constraints on some important issues related to the production of intra-plate continental basalts, such as the nature of mantle sources (lithosphere vs. asthenosphere) and origin of enriched components. Available research relies on the established cpx phenocryst-based approach. The uncertainty of this method is relatively large and this method cannot be applied to the basalts free of cpx phenocrysts. Thus, in the future, a new methodology with better accuracy and wide applicability should be developed. As we demonstrated, the partition coefficient of water between opx and cpx in peridotites shows a range of values. In the future, these values should be more precisely explored with an explanation for scattering and whether there is any relationship with geochemical and physical variables. Acknowledgments We thank Dr Zhong-He Zhou for inviting us to write this paper. Three anonymous referees are acknowledged for thorough reviews and providing constructive suggestions. FUNDING This work was supported by the National Natural Science Foundation of China (41225005). Conflict of interest statement. None declared. REFERENCES 1. Mackwell SJ , Kohlstedt DL , Paterson MS . The role of water in the deformation of olivine single-crystals . J Geophys Res 1985 ; 90 : 1319 – 33 . Google Scholar CrossRef Search ADS 2. Karato S , Paterson MS , Fitz-Gerald JD . Rheology of synthetic olivine aggregates: influence of grains size and water . J Geophys Res 1986 ; 91 : 8151 – 76 . Google Scholar CrossRef Search ADS 3. Karato S . The role of hydrogen in the electrical conductivity of the upper mantle . Nature 1990 ; 347 : 272 – 3 . Google Scholar CrossRef Search ADS 4. Graham CM , Elphick SC . Some experimental constraints on the role of hydrogen in oxygen and hydrogen diffusion and Al-Si interdiffusion in silicates . In: Ganguly J (ed.) . Diffusion, atomic ordering, and mass transport selected topics in geochemistry. Adv Phys Geochem 1991 ; 8 : 248 – 85 . Google Scholar CrossRef Search ADS 5. Meade C , Jeanloz R . Deep-focus earthquakes and recycling of water into the Earth's mantle . Science 1991 ; 252 : 68 – 72 . Google Scholar CrossRef Search ADS PubMed 6. Inoue T . Effect of water on melting phase relations and melt compositions in the system Mg2SiO4-MgSiO3-H2O up to 15 GPa . Phys Earth Planet Inter 1994 ; 85 : 237 – 63 . Google Scholar CrossRef Search ADS 7. Hirose K . Melting experiments on lherzolite KLB-1 under hydrous conditions and generation of high-magnesium andesitic melts . Geology 1997 ; 25 : 42 – 4 . Google Scholar CrossRef Search ADS 8. Mei S , Kohlstedt DL . Influence of water on plastic deformation of olivine aggregates. I: Diffusion creep regime . J Geophys Res 2000 ; 105 : 21457 – 69 . Google Scholar CrossRef Search ADS 9. Jung H , Karato S . Water-induced fabric transitions in olivine . Science 2001 ; 293 : 1460 – 3 . Google Scholar CrossRef Search ADS PubMed 10. Hofmeister AM . Enhancement of radiative transfer in the upper mantle by OH- in minerals . Phys Earth Planet Inter 2004 ; 146 : 483 – 95 . Google Scholar CrossRef Search ADS 11. Wang Z , Hiraga T , Kohlstedt DL . Effect of H+ on Fe-Mg interdiffusion in olivine, (Fe, Mg)2SiO4 . Appl Phys Lett 2004 ; 85 : 209 – 11 . Google Scholar CrossRef Search ADS 12. Hier-Majumder S , Mei S , Kohlstedt DL . Water weakening of clinopyroxeneite in diffusion creep . J Geophy Res 2005 ; 110 : B07406 . Google Scholar CrossRef Search ADS 13. Karato SI . Rheology of the deep upper mantle and its implications for the preservation of the continental roots: a review . Tectonophysics 2010 ; 481 : 82 – 98 . Google Scholar CrossRef Search ADS 14. Gaetani GA , Grove TL , Bryan WB . 1993 . The influence of water on the petrogenesis of subduction related igneous rocks . Nature 365 : 332 – 4 . Google Scholar CrossRef Search ADS 15. Hirose K , Kawamoto T . Hydrous partial melting of lherzolite at 1 GPa: The effect on H2O on the genesis of basaltic magmas . Earth Planet Sci Lett 1995 ; 133 : 463 – 73 . Google Scholar CrossRef Search ADS 16. Hirth G , Kohlstedt DL . Water in the oceanic upper mantle: Implications for rheology, melt extraction and the evolution of the lithosphere . Earth Planet Sci Lett 1996 ; 144 : 93 – 108 . Google Scholar CrossRef Search ADS 17. Asimow PD , Langmuir CH . 2003 . The important of water to oceanic mantle melting regimes . Nature 421 : 815 – 20 . Google Scholar CrossRef Search ADS PubMed 18. Dixon JE , Dixon TH , Bell DR et al. Lateral variation in upper mantle viscosity: role of water . Earth Planet Sci Lett 2004 ; 222 : 451 – 67 . Google Scholar CrossRef Search ADS 19. Hirschmann MM , Aubaud C , Withers AC . Storage capacity of H2O in nominally anhydrous minerals in the upper mantle . Earth Planet Sci Lett 2005 ; 236 : 167 – 81 . Google Scholar CrossRef Search ADS 20. Hirschmann MM , Tenner T , Aubaud C et al. Dehydration melting of nominally anhydrous mantle: the primacy of partitioning . Phys Earth Planet Inter 2009 ; 176 : 54 – 68 . Google Scholar CrossRef Search ADS 21. Hauri EH , Gaetani GA , Green TH . Partitioning of water during melting of the Earth's upper mantle at H2O-undersaturated conditions . Earth Planet Sci Lett 2006 ; 248 : 715 – 34 . Google Scholar CrossRef Search ADS 22. Li ZX , Lee CT , Peslier AH et al. Water contents in mantle xenoliths from the Colorado Plateau and vicinity: Implications for the mantle rheology and hydration-induced thinning of continental lithosphere . J Geophys Res 2008 ; 13 , B0921010.1029/2007jb005540 . 23. Grove TL , Till CB , Krawczynski MJ . The role of H2O in subduction zone magmatism . Annu Rev Earth Planet Sci 2012 ; 40 : 413 – 39 . Google Scholar CrossRef Search ADS 24. Green DH . Experimental petrology of peridotites, including effects of water and carbon on melting in the Earth's upper mantle . Phys Chem Minerals 2015 ; 42 : 95 – 122 . Google Scholar CrossRef Search ADS 25. Jordan TH . Composition and development of continental tectosphere . Nature 1978 ; 274 : 544 – 8 . Google Scholar CrossRef Search ADS 26. Pollack HN . Cratonization and thermal evolution of the mantle . Earth Planet Sci Lett 1986 ; 80 : 175 – 82 . Google Scholar CrossRef Search ADS 27. Xia QK , Liu J , Liu SC et al. High water content in Mesozoic primitive basalts of the North China Craton and implications on the destruction of cratonic mantle lithosphere . Earth Planet Sc Lett 2013 ; 361 : 85 – 97 . Google Scholar CrossRef Search ADS 28. Peslier AH , Woodland AB , Bell DR et al. Olivine water contents in the continental lithosphere and the longevity of cratons . Nature 2010 ; 467 : 78 – U108 . Google Scholar CrossRef Search ADS PubMed 29. Chen H , Xia QK , Ingrin J et al. Changing recycled oceanic components in the mantle source of the Shuangliao Cenozoic basalts, NE China: New constraints from water content . Tectonophysics . 2015 ; 650 : 113 – 23 . Google Scholar CrossRef Search ADS 30. Liu J , Xia QK , Deloule E et al. Recycled oceanic crust and marine sediment in the source of alkali basalts in Shandong, eastern China: evidence from magma water content and oxygen isotopes . J Geophys Res 2015 ; 120 : 8281 – 303 . Google Scholar CrossRef Search ADS 31. Ohtani E . Hydrous minerals and the storage of water in the deep mantle . Chem Geol 2015 ; 418 : 6 – 15 . Google Scholar CrossRef Search ADS 32. Peslier AH . A review of water contents of nominally anhydrous minerals in the mantles of Earth, Mars and the Moon . J Vol Geothem Res 2010 ; 197 : 239 – 58 . Google Scholar CrossRef Search ADS 33. Demouchy S , Bolfan-Casanova N . Distribution and transport of hydrogen in the lithospheric mantle: A Review . Lithos 2016 ; 240-243 : 402 – 25 . Google Scholar CrossRef Search ADS 34. Bell DR , Rossman GR . Water in Earth's mantle: the role of nominally anhydrous minerals . Science 1992 ; 255 : 1391 – 7 . Google Scholar CrossRef Search ADS PubMed 35. Rossman GR . Analytical methods for measuring water in nominally anhydrous minerals . Rev Mineral Geochem 2006 ; 62 : 1 – 28 . Google Scholar CrossRef Search ADS 36. Aubaud C , Hauri EH , Hirschmann MM . Hydrogen partition coefficients between nominally anhydrous minerals and basaltic melts . Geophys Res Lett 2004 ; 31 : 2 – 5 . Google Scholar CrossRef Search ADS 37. Yang XZ , Deloule E , Xia QK et al. Water contrast between Precambrian and Phanerozoic continental lower crust in eastern China . J Geophys Res 2008 ; 113 : B08207 . Google Scholar CrossRef Search ADS 38. Kovács I , Hermann J , O’Neill HSC et al. Quantitative absorbance spectroscopy with unpolarized light: Part II. Experimental evaluation and development of a protocol for quantitative analysis of mineral IR spectra . Am Mineral 2008 ; 93 : 765 – 78 . Google Scholar CrossRef Search ADS 39. Aubaud C , Withers AC , Hirschmann MM et al. Intercalibration of FTIR and SIMS for hydrogen measurements in glasses and nominally anhydrous minerals . Am Mineral 2007 ; 92 : 811 – 28 . Google Scholar CrossRef Search ADS 40. Xia QK , Hao YT , Li P et al. Low water content of the Cenozoic lithospheric mantle beneath the eastern part of the North China Craton . J Geophy Res 2010 ; 115 : B07207 . Google Scholar CrossRef Search ADS 41. Xia QK , Hao YT , Liu SC et al. Water contents of the Cenozoic lithospheric mantle beneath the western part of the North China Craton: Peridotite xenolith constraints . Gond Res 2013 ; 23 : 108 – 18 . Google Scholar CrossRef Search ADS 42. Yu Y , Xu XS , Griffin WL et al. H2O contents and their modification in the Cenozoic subcontinental lithospheric mantle beneath the Cathaysia block, SE China . Lithos 2011 ; 126 : 182 – 97 . Google Scholar CrossRef Search ADS 43. Hao YT , Xia QK , Liu SC et al. Recognizing juvenile and relict lithospheric mantle beneath the North China Craton: combined analysis of H2O major and trace elements and Sr-Nd isotope compositions of clinopyroxenes . Lithos 2012 ; 149 : 136 – 45 . Google Scholar CrossRef Search ADS 44. Hao YT , Xia QK , Li QW et al. Partial melting control of water contents in the Cenozoic lithospheric mantle of the Cathaysia block of South China . Chem Geol 2014 ; 380 : 7 – 19 . Google Scholar CrossRef Search ADS 45. Hao YT , Xia QK , Jia ZB et al. Regional heterogeneity in the water content of the Cenozoic lithospheric mantle of Eastern China . J Geophys Res 2016 ; 121 : 1 – 21 . 46. Wang Q , Bagdassarov N , Xia QK et al. Water contents and electrical conductivity of peridotite xenoliths from the North China Craton: implications for water distribution in the upper mantle . Lithos 2014 ; 189 : 105 – 26 . Google Scholar CrossRef Search ADS 47. Li P , Xia QK , Deloule E et al. Temporal variation of H2O content in the lithospheric mantle beneath the eastern North China Craton: implications for the destruction of cratons . Gond Res 2015 ; 28 : 276 – 87 . Google Scholar CrossRef Search ADS 48. Demouchy S , Ishikawa A , Tommasi A et al. Characterization of hydration in the mantle lithosphere: peridotite xenoliths from the Ontong Java Plateau as an example . Lithos 2015 ; 212-215 : 189 – 201 . Google Scholar CrossRef Search ADS 49. Denis CMM , Alard O , Demouchy S . Water content and hydrogen behaviour during metasomatism in the uppermost mantle beneath Ray Pic volcano (Massif Central, France) . Lithos 2015 ; 236–7 : 256 – 74 . Google Scholar CrossRef Search ADS 50. Grant K , Ingrin J , Lorand JP et al. Water partitioning between mantle minerals from peridotite xenoliths . Contrib Mineral Petrol 2007 ; 154 : 15 – 34 . Google Scholar CrossRef Search ADS 51. Ingrin J , Skogby H . Hydrogen in nominally anhydrous upper-mantle minerals: concentration levels and implications . Eur J Mineral 2000 ; 12 : 543 – 70 . Google Scholar CrossRef Search ADS 52. Peslier AH , Luhr JF , Post J . Low water contents in pyroxenes from spinel-peridotites of the oxidized, sub-arc mantle wedge . Earth Planet Sci Lett 2002 ; 201 : 69 – 86 . Google Scholar CrossRef Search ADS 53. Peslier AH , Woodland AB , Bell DR et al. Metasomatic control of water contents in the Kaapvaal cratonic mantle . Geochim Cosmochim Acta 2012 ; 97 : 213 – 46 . Google Scholar CrossRef Search ADS 54. Skogby H , Bell DR , Rossman GR . Hydroxide in pyroxene: variation in the natural enviroment . Am Mineral 1990 ; 75 : 764 – 74 . 55. Sundvall R , Stalder R . Water in upper mantle pyroxene megacrysts and xenocrysts: a survey study . Am Mineral 2011 ; 96 : 1215 – 27 . Google Scholar CrossRef Search ADS 56. Warren J , Hauri E . Pyroxenes as tracers of mantle water variations . J Geophy Res 2014 ; 119 : 1851 – 81 . Google Scholar CrossRef Search ADS 57. Tian ZZ , Liu J , Xia QK et al. Water concentration profiles in natural mantle orthopyroxenes: a geochronometer for long annealing of xenoliths within magma . Geology 2017 ; 45 : 87 – 90 . Google Scholar CrossRef Search ADS 58. Yang XZ , Xia QK , Deloule E et al. Water in minerals of continental lithospheric mantle and overlying lower crust: a comparative study of peridotite and granulite xenoliths from the North China Craton . Chem Geol 2008 ; 256 : 33 – 45 . Google Scholar CrossRef Search ADS 59. Koga K , Hauri E , Hirschmann M et al. Hydrogen concentration analyses using SIMS and FTIR: comparison and calibration for nominally anhydrous minerals . Geochem Geophys Geosyst 2003 ; 4 , 10.1029/2002GC000378 . 60. Withers AC , Hirschmann MM . H2O storage capacity of MgSiO3 clinoenstatite at 8-13 GPa, 1,100-1,400°C . Contrib Mineral Petrol 2007 ; 154 : 663 – 74 . Google Scholar CrossRef Search ADS 61. Withers AC , Hirschmann MM . Influence of temperature, composition, silica activity and oxygen fugacity on the H2O storage capacity of olivine at 8 GPa . Contrib Mineral Petrol 2008 ; 156 : 595 – 605 . Google Scholar CrossRef Search ADS 62. Tenner TJ , Hirschmann MM , Withers AC et al. Hydrogen partitioning between nominally anhydrous upper mantle minerals and melt between 3 and 5 GPa and applications to hydrous peridotite partial melting . Chem Geol 2009 ; 262 : 42 – 56 . Google Scholar CrossRef Search ADS 63. Adam J , Turner M , Hauri EH et al. Crystal/melt partitioning of water and other volatiles during the near-solidus melting of mantle peridotite: comparisons with non-volatile incompatible elements and implications for the generation of intraplate magmatism . Am Mineral 2016 ; 101 : 876 – 88 . Google Scholar CrossRef Search ADS 64. Asimow PD , Dixon JE , Langmuir CH . A hydrous melting and fractionation model for mid-ocean ridge basalts: application to the Mid-Atlantic Ridge near the Azores . Geochem Geophy Geosyst 2004 ; 5 , 10.1029/2003GC000568 . 65. Dixon JE , Stolper E , Delaney JR . Infrared spectroscopic measurements of CO2 and H2O in Juan de Fuca Ridge basaltic glasses . Earth Planet Sci Lett 1988 ; 90 : 87 – 104 . Google Scholar CrossRef Search ADS 66. Michael PJ . The concentration, behavior and storage of H2O in the suboceanic upper mantle: implications for mantle metasomatism . Geochim Cosmochim Acta 1988 ; 52 : 555 – 66 . Google Scholar CrossRef Search ADS 67. Saal AE , Hauri EH , Langmuir CH et al. Vapour undersaturation in primitive mid-ocean-ridge basalt and the volatile content of Earth's upper mantle . Nature 2002 ; 419 : 451 – 5 . Google Scholar CrossRef Search ADS PubMed 68. Simons K . Volatiles in basaltic glasses from the Easter-Salas y Gomez Seamount Chain and Easter Microplate: implications for geochemical cycling of volatile elements . Geochem Geophys Geosyst 2002 ; 3 , 10.1029/2001GC000173 . 69. Sobolev AV , Chaussidon M . H2O concentrations in primary melts from supra-subduction zones and mid-ocean ridges: implications for H2O storage and recycling in the mantle . Earth Planet Sci Lett 1996 ; 137 : 45 – 55 . Google Scholar CrossRef Search ADS 70. Dixon JE , Clague DA , Wallace P et al. Volatiles in alkalic basalts from the North Arch volcanic field, Hawaii: extensive degassing of deep submarine-erupted alkalic series lavas . J Petrol 1997 ; 38 : 911 – 39 . Google Scholar CrossRef Search ADS 71. Nichols ARL , Carroll MR , Höskuldsson Á . Is the Iceland hot spot also wet? Evidence from the water contents of undegassed submarine and subglacial pillow basalts . Earth Planet Sci Lett 2002 ; 202 : 77 – 87 . Google Scholar CrossRef Search ADS 72. Seaman C , Sherman SB , Garcia MO et al. Volatiles in glasses from the HSDP2 drill core . Geochem Geophys Geosystems 2004 ; 5 , 10.1029/2003GC000596 . 73. Wallace PJ . Volatiles in submarine basaltic glasses from the Northern Kerguelen Plateau (ODP Site 1140): implications for source region compositions, magmatic processes, and plateau subsidence . J Petrol 2002 ; 43 : 1311 – 26 . Google Scholar CrossRef Search ADS 74. Workman RK , Hauri E , Hart SR et al. Volatile and trace elements in basaltic glasses from Samoa: implications for water distribution in the mantle . Earth Planet Sci Lett 2006 ; 241 : 932 – 51 . Google Scholar CrossRef Search ADS 75. Falus G , Tommasi A , Ingrin J et al. Deformation and seismic anisotropy of the lithospheric mantle in the southeastern Carpathians inferred from the study of mantle xenoliths . Earth Planet Sci Lett 2008 ; 272 : 50 – 64 . Google Scholar CrossRef Search ADS 76. Kovács I , Green DH , Rosenthal A et al. An experimental study of water in nominally anhydrous minerals in the upper mantle near the water-saturated solidus . J Petrol 2012 ; 53 : 2067 – 93 . Google Scholar CrossRef Search ADS 77. Novella D , Frost DJ . The composition of hydrous partial melts of garnet peridotite at 6 GPa: implications for the origin of Group II Kimberlites . J Petrol 2014 ; 55 : 2097 – 124 . Google Scholar CrossRef Search ADS 78. Hao YT , Xia QK , Tian ZZ et al. Mantle metasomatism did not modify the water content of the peridotite xenoliths from the Tianchang basalts of eastern China . Lithos 2016 ; 260 : 315 – 27 . Google Scholar CrossRef Search ADS 79. Liu SC , Xia QK . Water content in the early Cretaceous lithospheric mantle beneath the south-central Taihang Mountains: implications for the destruction of the North China Craton . Chin Sci Bull 2014 ; 59 : 1362 – 5 . Google Scholar CrossRef Search ADS 80. Peslier AH , Woodland AB , Wolff JA . Fast kimberlite ascent rates estimated from hydrogen diffusion profiles in xenolithic mantle olivines from southern Africa . Geochim Cosmochim Acta 2008 ; 72 : 2711 – 22 . Google Scholar CrossRef Search ADS 81. Baptiste V , Tommasi A , Demouchy S . Deformation and hydration of the lithospheric mantle beneath the Kaapvaal craton, South Africa . Lithos 2012 ; 149 : 31 – 50 . Google Scholar CrossRef Search ADS 82. Xia QK , Yang X , Deloule E et al. Water in the lower crustal granulite xenoliths from Nushan, eastern China . J Geophys Res 2006 ; B11202 , 10.1029/2006JB004296 . 83. Németh B , Török K , Kovács I et al. Melting, fluid migration and fluid-rock interactions in the lower crust beneath the Bakony-Balaton Highland volcanic field: a silicate melt and fluid inclusion study . Miner Petrol 2015 ; 109 : 217 – 34 . Google Scholar CrossRef Search ADS 84. Rudnick RL , Gao S . Composition of the continental crust . In: Rudnick RL (ed.) . Treatise on Geochemistry: The Crust . Oxford: Elsevier-Pergamon , 2003 , 1 – 64 . 85. Yang XZ . An experimental study of H solubility in feldspars: effect of composition, oxygen fugacity, temperature and pressure and implications for crustal processes . Geochim Cosmochim Acta 2012 ; 97 : 46 – 57 . Google Scholar CrossRef Search ADS 86. Keppler H , Bolfan-Casanova N . Thermodynamics of water solubility and partitioning . Rev Mineral Geochem 2006 ; 62 : 193 – 230 . Google Scholar CrossRef Search ADS 87. Yang XZ , Gaillard F , Scaillet B . A relatively reduced Hadean continental crust and implications for the early atmosphere and crustal rheology . Earth Planet Sci Lett 2014 ; 393 : 210 – 9 . Google Scholar CrossRef Search ADS 88. Tang YJ , Zhang HF , Ying JF . Asthenosphere-litho spheric mantle interaction in an extensional regime: implication from the geochemistry of Cenozoic basalts from Taihang Mountains, North China Craton . Chem Geol 2006 ; 233 : 309 – 27 . Google Scholar CrossRef Search ADS 89. Zeng G , Chen LH , Hofmann AW et al. Crust recycling in the sources of two parallel volcanic chains in Shandong, North China . Earth Planet Sci Lett 2011 ; 302 : 359 – 68 . Google Scholar CrossRef Search ADS 90. Chen LH , Zeng G , Jiang SY et al. Sources of Anfengshan basalts: subducted lower crust in the Sulu UHP belt, China . Earth Planet Sc Lett . 2009 ; 286 : 426 – 35 . Google Scholar CrossRef Search ADS 91. Zhang JJ , Zheng YF , Zhao ZF . Geochemical evidence for interaction between oceanic crust and lithospheric mantle in the origin of Cenozoic continental basalts in east-central China . Lithos 2009 ; 110 : 305 – 26 . Google Scholar CrossRef Search ADS 92. Xu Z , Zhao ZF , Zheng YF . Slab-mantle interaction for thinning of cratonic lithospheric mantle in North China: geochemical evidence from Cenozoic continental basalts in central Shandong . Lithos 2012 ; 146 : 202 – 17 . Google Scholar CrossRef Search ADS 93. Xu YG , Zhang HH , Qiu HN et al. Oceanic crust components in continental basalts from Shuangliao, Northeast China: derived from the mantle transition zone? Chem Geol 2012 ; 328 : 168 – 84 . Google Scholar CrossRef Search ADS 94. Kuritani T , Ohtani E , Kimura JI . Intensive hydration of the mantle transition zone beneath China caused by ancient slab stagnation . Nat Geosci 2011 ; 4 : 713 – 6 . Google Scholar CrossRef Search ADS 95. Kuritani T , Kimura JI , Ohtani E et al. Transition zone origin of potassic basalts from Wudalianchi volcano, northeast China . Lithos 2013 ; 156 : 1 – 12 . Google Scholar CrossRef Search ADS 96. Michael PJ . Regionally distinctive sources of depleted MORB: evidence from trace-elements and H2O . Earth Planet Sc Lett 1995 ; 131 : 301 – 20 . Google Scholar CrossRef Search ADS 97. Dixon JE , Clague DA . Volatiles in basaltic glasses from Loihi seamount, Hawaii: evidence for a relatively dry plume component . J Petrol 2001 ; 42 : 627 – 54 . Google Scholar CrossRef Search ADS 98. Dixon JE , Leist L , Langmuir C et al. Recycled dehydrated lithosphere observed in plume-influenced mid-ocean-ridge basalt . Nature 2002 ; 420 : 385 – 9 . Google Scholar CrossRef Search ADS PubMed 99. Chen Y , Zhang Y , Graham D et al. Geochemistry of Cenozoic basalts and mantle xenoliths in Northeast China . Lithos 2007 ; 96 : 108 – 26 . Google Scholar CrossRef Search ADS 100. Hong LB , Zhang YH , Qian SP et al. Constraints from melt inclusions and their host olivines on the petrogenesis of Oligocene-Early Miocene Xindian basalts, Chifeng area, North China Craton . Contrib Mineral Petrol 2013 ; 165 : 305 – 26 . Google Scholar CrossRef Search ADS 101. Wang Y , Zhao ZF , Zheng YF et al. Geochemical constraints on the nature of mantle source for Cenozoic continental basalts in east-central China . Lithos 2011 ; 125 : 940 – 55 . Google Scholar CrossRef Search ADS 102. Chen H , Xia QK , Ingrin J . Water content of the Xiaogulihe ultrapotassic volcanic rocks, NE China: implications for the source of the potassium-rich component . Sci Bull . 2015 ; 60 : 1468 – 70 . Google Scholar CrossRef Search ADS 103. Chen H , Xia QK , Ingrin J et al. Heterogeneous source components of intraplate basalts from NE China induced by the ongoing Pacific slab subduction . Earth Planet Sci Lett 2017 ; 459 : 208 – 20 . Google Scholar CrossRef Search ADS 104. Liu J , Xia QK , Deloule E et al. Water content and oxygen isotopic composition of alkali basalts from the Taihang Mountains, China: recycled oceanic components in the mantle source . J Petrol 2015 ; 56 : 681 – 702 . Google Scholar CrossRef Search ADS 105. Liu SC , Xia QK , Choi SH et al. Continuous supply of recycled Pacific oceanic materials in the source of Cenozoic basalts in SE China: the Zhejiang case . Contrib Mineral Petrol 2016 ; 171 : 1 – 31 . Google Scholar CrossRef Search ADS 106. Wade JA , Plank T , Hauri EH et al. Prediction of magmatic water contents via measurement of H2O in clinopyroxene phenocrysts . Geology 2008 ; 36 : 799 – 802 . Google Scholar CrossRef Search ADS 107. Danyushevsky LV , Falloon TJ , Sobolev AV et al. The H2O content of basalt glasses from Southwest Pacific back-arc basins . Earth Planet Sci Lett 1993 ; 117 : 347 – 62 . Google Scholar CrossRef Search ADS 108. Danyushevsky LV , Eggins SM , Falloon TJ et al. H2O abundance in depleted tomoderately enrichedmid-ocean ridge magmas; part I: incompatible behaviour, implications for mantle storage, and origin of regional variations . J Petrol 2000 ; 41 : 1329 – 64 . Google Scholar CrossRef Search ADS 109. Dobson PF , Skogby H , Rossman GR . Water in boninite glass and coexisting orthopyroxene: concentration and partitioning . Contrib Mineral Petrol 1995 ; 118 : 414 – 9 . Google Scholar CrossRef Search ADS 110. Hochstaedter AG , Gill JB , Kusakabe M et al. Volcanism in the Sumisu Rift. I. Major element, volatile, and stable isotope geochemistry . Earth Planet Sci Lett 1990 ; 100 : 179 – 94 . Google Scholar CrossRef Search ADS 111. Sisson T , Layne G . H2O in basalt and basaltic andesite glass inclusions from four subduction-related volcanoes . Earth Planet Sci Lett 1993 ; 117 : 619 – 35 . Google Scholar CrossRef Search ADS 112. Stolper E , Newman S . The role of water in the petrogenesis of Mariana Trough magmas . Earth Planet Sci Lett 1994 ; 121 : 293 – 325 . Google Scholar CrossRef Search ADS 113. Wallace PJ . Water and partial melting in mantle plumes: inferences from the dissolved H2O concentrations of Hawaiian basaltic magmas . Geophys Res Lett 1998 ; 25 : 3639 – 42 . Google Scholar CrossRef Search ADS 114. Wallace PJ . Volatiles in subduction zonemagmas: concentrations and fluxes based on melt inclusion and volcanic gas data . J Volcanol Geotherm Res 2005 ; 140 : 217 – 40 . Google Scholar CrossRef Search ADS 115. Sun SS , McDonough WF . Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes . Geological Society, London, Special Publications 1989 ; 42 : 313 – 45 . Google Scholar CrossRef Search ADS 116. Cabral RA , Jackson MG , Koga KT et al. Volatile cycling of H2O, CO2, F, and Cl in the HIMU mantle: a new window provided by melt inclusions from oceanic hot spot lavas at Mangaia, Cook Islands . Geochem Geophys Geosyst 2014 ; 15 : 4445 – 67 . Google Scholar CrossRef Search ADS 117. Kendrick MA , Jackson MG , Kent AJ et al. Contrasting behaviours of CO2, S, H2O and halogens (F, Cl, Br, and I) in enriched-mantle melts from Pitcairn and Society seamounts . Chem Geol 2014 ; 370 : 69 – 81 . Google Scholar CrossRef Search ADS 118. Kendrick MA , Jackson MG , Hauri EH et al. The halogen (F, Cl, Br, I) and H2O systematics of Samoan lavas: assimilated-seawater, EM2 and high-3He/4He components . Earth Planet Sci Lett 2015 ; 410 : 197 – 209 . Google Scholar CrossRef Search ADS 119. Ruscitto DM , Wallace PJ , Cooper LB et al. Global variations in H2O/Ce. 2. Relationships to arc magma geochemistry and volatile fluxes . Geochem Geophys Geosyst 2012 ; 13 , 10.1029/2011GC003887 . 120. Wang XC , Wilde SA , Li QL et al. Continental flood basalts derived from the hydrous mantle transition zone . Nat Commun 2015 ; 6 : 7700 , 10.1038/ncomms8700 . Google Scholar CrossRef Search ADS PubMed 121. Chen H , Xia QK , Liu J et al. MTZ source of the Cenozoic basalts of eastern China constrained by H2O content . 2017 ; in preparation . 122. Workman RK , Hart SR . Major and trace element composition of the depleted MORB mantle (DMM) . Earth Planet Sc Lett 2005 ; 231 : 53 – 72 . Google Scholar CrossRef Search ADS 123. Zhi X , Song Y , Frey FA et al. Geochemistry of Hannuoba basalts, eastern China: constraints on the origin of continental alkalic and tholeiitic basalt . Chem Geol 1990 ; 88 : 1 – 33 . Google Scholar CrossRef Search ADS 124. Xu YG , Ma JL , Frey FA et al. Role of lithosphere–asthenosphere interaction in the genesis of Quaternary alkali and tholeiitic basalts from Datong, western North China Craton . Chem Geol 2005 ; 224 : 247 – 71 . Google Scholar CrossRef Search ADS 125. Sakuyama T , Tian W , Kimura JI et al. Melting of dehydrated oceanic crust from the stagnant slab and of the hydrated mantle transition zone: constraints from Cenozoic alkaline basalts in eastern China . Chem Geol 2013 ; 359 : 32 – 48 . Google Scholar CrossRef Search ADS 126. Xu YG . Recycled oceanic crust in the source of 90–40 Ma basalts in North and Northeast China: evidence, provenance and significance . Geochim Cosmochim Acta 2014 ; 143 : 49 – 67 . Google Scholar CrossRef Search ADS 127. Xu Z , Zheng YF , Zhao ZF et al. The hydrous properties of subcontinental lithospheric mantle: constraints from water content and hydrogen isotope composition of phenocrysts from Cenozoic continental basalt in North China . Geochim Cosmochim Acta 2014 ; 143 : 285 – 302 . Google Scholar CrossRef Search ADS 128. Bodnar RJ , Azbej T , Becker SP et al. Whole Earth geohydrologic cycle, from the clouds to the core: the distribution of water in the dynamic Earth system . In: Bickford ME (ed.). The web of geological sciences: advances, impacts, and interactions . Geol Soc Am Spec Paper 2013 ; 500 : 431 – 61 . Google Scholar CrossRef Search ADS 129. Panero WR , Pigott JS , Reaman DM et al. Dry (Mg, Fe)SiO3 perovskite in the Earth's lower mantle . J Geophys Res 2015 ; 120 : 894 – 908 . Google Scholar CrossRef Search ADS 130. Pearson DG , Brenker FE , Nestola F et al. Hydrous mantle transition zone indicated by ringwoodite included within diamond . Nature 2014 ; 507 : 221 – 4 . Google Scholar CrossRef Search ADS PubMed 131. Fukao Y , Obayashi M , Inoue H et al. Subducting slabs stagnant in the mantle transition zone . J Geophys Res 1992 ; 97 : 4809 – 22 . Google Scholar CrossRef Search ADS 132. Kelbert A , Schultz A , Egbert G . Global electromagnetic induction constraints on transition-zone water content variations . Nature 2009 ; 460 : 1003 – 6 . Google Scholar CrossRef Search ADS PubMed 133. Zhang J , Zhang HF , Ying JF et al. Contribution of subducted Pacific slab to Late Cretaceous mafic magmatism in Qingdao region, China: a petrological record . Isl Arc 2008 ; 17 : 231 – 41 . Google Scholar CrossRef Search ADS 134. Yang W , Teng FZ , Zhang HF et al. Magnesium isotopic systematics of continental basalts from the North China craton: implications for tracing subducted carbonate in the mantle . Chem Geol 2012 ; 328 : 185 – 94 . Google Scholar CrossRef Search ADS 135. Huang JL , Zhao DP . High-resolution mantle tomography of China and surrounding regions . J Geophys Res 2006 ; 111 ( B9 ), 10.1029/2005JB004066 . 136. Kuang YS , Wei X , Hong LB et al. Petrogenetic evaluation of the Laohutai basalts from North China Craton: melting of a two-component source during lithospheric thinning in the late Cretaceous-early Cenozoic . Lithos 2012 ; 154 : 68 – 82 . Google Scholar CrossRef Search ADS 137. Li HY , Huang XL , Guo H . Geochemistry of Cenozoic basalts from the Bohai Bay Basin: implications for a heterogeneous mantle source and lithospheric evolution beneath the eastern North China Craton . Lithos 2014 ; 196 : 54 – 66 . Google Scholar CrossRef Search ADS 138. Li HY , Xu YG , Ryan JG et al. Olivine and melt inclusion chemical constraints on the source of intracontinental basalts from the eastern North China Craton: discrimination of contributions from the subducted Pacific slab . Geochim Cosmochim Acta 2016 ; 178 : 1 – 19 . Google Scholar CrossRef Search ADS 139. Hawkesworth CJ , Hergt JM , Ellam RM et al. Element fluxes associated with subduction related magmatism . Philosophical Transactions of the Royal Society of London Series a-Mathematical Physical and Engineering Sciences 1991 ; 335 : 393 – 405 . Google Scholar CrossRef Search ADS 140. Tatsumi Y , Eggins S . Subduction Zone Magmatism . Oxford: Blackwell Science , 1995 , 211 . 141. Zheng YF . Fluid regime in continental subduction zones: petrological insights from ultrahigh-pressure metamorphic rocks . J Geol Soc 2009 ; 166 : 763 – 82 . Google Scholar CrossRef Search ADS 142. Zheng YF , Xia QX , Chen RX et al. Partial melting, fluid supercriticality and element mobility in ultrahigh-pressure metamorphic rocks during continental collision . Earth Sci Rev 2011 ; 107 : 342 – 74 . Google Scholar CrossRef Search ADS 143. Zhang JF , Jin ZM , Green HW et al. Hydroxyl in continental deep subduction zone: evidence from UHP eclogites of the Dabie Mountains . Chin Sci Bull 2001 ; 46 : 592 – 6 . Google Scholar CrossRef Search ADS 144. Zhang JF , Green HW , Bozhilov K et al. Faulting induced by precipitation of water at grain boundaries in hot subducting oceanic crust . Nature 2004 ; 428 : 633 – 6 . Google Scholar CrossRef Search ADS PubMed 145. Xia QK , Sheng YM , Yang XZ et al. Heterogeneity of water in garnets from UHP eclogites, eastern Dabieshan, China . Chem Geol 2005 ; 224 : 237 – 46 . Google Scholar CrossRef Search ADS 146. Sheng Y M , Xia Q K , Dallai L et al. H2O contents and D/H ratios of nominally anhydrous minerals from ultrahigh-pressure eclogites of the Dabie orogen, eastern China . Geochim Cosmochim Acta 2007 ; 71 : 2079 – 103 . Google Scholar CrossRef Search ADS 147. Chen RX , Zheng YF , Gong B et al. Origin of retrograde fluid in ultrahigh-pressure metamorphic rocks: constraints from mineral hydrogen isotope and water content changes in eclogite-gneiss transitions in the Sulu orogen . Geochim Cosmochim Acta 2007 ; 71 : 2299 – 325 . Google Scholar CrossRef Search ADS 148. Zhao ZF , Chen B , Zheng YF et al. Mineral oxygen isotope and hydroxyl content changes in ultrahigh-pressure eclogite-gneiss contacts from Chinese Continental Scientific Drilling Project cores . J Metamorph Geol 2007 ; 25 : 165 – 86 . Google Scholar CrossRef Search ADS 149. Zhao XD , Zhang ZM . Water in omphacites from UHP eclogites in the CCSD main hole . Acta Petrol Sin 2007 ; 22 : 2039 – 50 . 150. Aines , RD , Rossman GR . Water content of mantle garnets . Geology 1984 ; 12 : 720 – 3 . Google Scholar CrossRef Search ADS 151. Langer K , Rovarick E , Sobolev NV et al. Single-crystal spectra of garnets from diamondiferous high-pressure metamorphic rocks from Kazakhstan: indications for OH–, H2O, and FeTi charge transfer . Eur J Mineral 1993 ; 5 : 1091 – 100 . Google Scholar CrossRef Search ADS 152. Snyder GA , Taylor LA , Jerde EA et al. Archean mantle heterogeneity and the origin of diamondiferous eclogites, Siberia: evidence from stable isotopes and hydroxyl in garnet . Am Mineral 1995 ; 80 : 799 – 809 . Google Scholar CrossRef Search ADS 153. Matsyuk SS , Langer K , Hosch A . Hydroxyl defects in garnets from mantle xenoliths in kimberlites of the Siberian platform . Contrib Mineral Petrol 1998 ; 132 : 163 – 79 . Google Scholar CrossRef Search ADS 154. Ragozin AL , Karimovac AA , Litasov KD et al. Water content in minerals of mantle xenoliths from the Udachnaya pipe kimberlites (Yakutia) . Russ Geol Geophy 2014 ; 55 : 428 – 42 . Google Scholar CrossRef Search ADS 155. Bell DR , Rossman GR , Moore RO . Abundance and partitioning of OH in a high-pressure magmatic system: megacrysts from the Monastery kimberlite, South Africa . J Petrol 2004 ; 45 : 1539 – 64 . Google Scholar CrossRef Search ADS 156. Katayama I , Nakashima S , Yurimoto H . Water content in natural eclogite and implication for water transport into the deep upper mantle . Lithos 2006 ; 86 : 245 – 59 . Google Scholar CrossRef Search ADS 157. Skogby H , Janák M , Broska I . Water incorporation in omphacite: concentrations and compositional relations in ultrahigh-pressure eclogites from Pohorje, Eastern Alps . Eur J Mineral 2016 ; 28 : 631 – 9 . Google Scholar CrossRef Search ADS 158. Su W , You ZD , Cong BL et al. Cluster of water molecules in garnet from ultrahigh-pressure eclogite . Geology 2002 ; 30 : 611 – 4 . Google Scholar CrossRef Search ADS 159. Liu FL , Xu ZQ . Fluid inclusions hidden in coesite-bearing zircons in ultrahigh-pressure metamorphic rocks from southwestern Sulu terrane in eastern China . Chin Sci Bull 2004 ; 49 : 396 – 404 . Google Scholar CrossRef Search ADS 160. Zhang ZM , Shen K , Xiao YL et al. Mineral and fluid inclusions in zircon of UHP metamorphic rocks from the CCSD-main drill hole: a record of metamorphism and fluid activity . Lithos 2006 ; 92 : 378 – 98 . Google Scholar CrossRef Search ADS 161. Meng DW , Wu XL , Fan XY et al. Submicron-sized fluid inclusions and distribution of hydrous components in jadeite, quartz and symplectite-forming minerals from UHP jadeite-quartzite in the Dabie Mountains, China: TEM and FTIR investigation . App Geochem 2009 ; 24 : 517 – 26 . Google Scholar CrossRef Search ADS 162. Xiao YL , Zhang ZM , Hoefs J et al. Ultrahigh-pressure metamorphic rocks from the Chinese Continental Scientific Drilling Project. II. Oxygen isotope and fluid inclusion distributions through vertical sections . Contrib Mineral Petrol 2006 ; 152 : 443 – 58 . Google Scholar CrossRef Search ADS 163. Zheng Y , Zhao Z , Chen Y . Continental subduction channel processes: Plate interface interaction during continental collision . Chin Sci Bull 2013 ; 58 : 4371 – 7 . Google Scholar CrossRef Search ADS 164. Katayama I , Nakashima S . Hydroxyl in clinopyroxene from the deep subducted crust: evidence for H2O transport into the mantle . Am Mineral 2003 ; 88 : 229 – 34 . Google Scholar CrossRef Search ADS 165. Su W , Ji ZP , Ye K et al. Distribution of hydrous components in jadeite of the Dabie Mountains . Earth Planet Sci Lett 2004 ; 222 : 85 – 100 . Google Scholar CrossRef Search ADS 166. Hirschmann MM . Water, melting and the deep Earth water cycle . Annu Rev Earth Planet Sci 2006 ; 34 : 629 – 53 . Google Scholar CrossRef Search ADS 167. Li XP , Zheng YF , Wu YB et al. Low-T eclogite in the Dabie terrane of China: petrological and isotopic constraints on fluid activity and radiometric dating . Contrib Mineral Petrol 2004 ; 148 : 443 – 70 . Google Scholar CrossRef Search ADS 168. Guo S , Ye K , Chen Y et al. Fluid–rock interaction and element mobilization in UHP metabasalt: constraints from an omphacite–epidote vein and host eclogites in the Dabie orogen . Lithos 2012 ; 136–9 : 145 – 67 . Google Scholar CrossRef Search ADS 169. Lu R , Keppler H . Water solubility in pyrope to l00 kbar . Contrib Mineral Petrol 1997 ; 129 : 35 – 42 . Google Scholar CrossRef Search ADS 170. Withers AC , Wood BJ , Carroll MR . The OH contents of pyrope at high pressure . Chem Geol 1998 ; 147 : 161 – 71 . Google Scholar CrossRef Search ADS 171. Mosenfelder JL . Pressure dependence of hydroxyl solubility in coesite . Phys Chem Miner 2000 ; 27 : 610 – 7 . Google Scholar CrossRef Search ADS 172. Rauch M , Keppler H . Water solubility in orthopyroxene . Contrib Mineral Petrol 2002 ; 143 : 525 – 36 . Google Scholar CrossRef Search ADS 173. Koch-Müller M , Dera P , Fei YW et al. OH- in synthetic and natural coesite . Am Mineral 2003 ; 88 : 1436 – 45 . Google Scholar CrossRef Search ADS 174. Bromiley G . Solubility of hydrogen and ferric iron in rutile and TiO2 (II): implications for phase assemblages during ultrahigh-pressure metamorphism and for the stability of silica polymorphs in the lower mantle . Geophys Res Lett 2004 ; 31 : L04610 , 10.1029/2004GL019430(4) . Google Scholar CrossRef Search ADS 175. Mierdel K , Keppler H . The temperature dependence of water solubility in enstatite . Contrib Mineral Petrol 2004 ; 148 : 305 – 11 . Google Scholar CrossRef Search ADS 176. Yang XZ , McCammon C . Fe3+-rich augite and high electrical conductivity in the deep lithosphere . Geology 2012 ; 40 : 131 – 4 . Google Scholar CrossRef Search ADS 177. Bizimis M , Peslier AH . Water in Hawaiian garnet pyroxenites: implications for water heterogeneity in the mantle . Chem Geol 2015 ; 397 : 61 – 75 . Google Scholar CrossRef Search ADS © The Author(s) 2017. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png National Science Review Oxford University Press

# Water in the upper mantle and deep crust of eastern China: concentration, distribution and implications

, Volume Advance Article – Feb 24, 2017
20 pages

/lp/ou_press/water-in-the-upper-mantle-and-deep-crust-of-eastern-china-8ebVJ0S6Rw
Publisher
Oxford University Press
ISSN
2095-5138
eISSN
2053-714X
D.O.I.
10.1093/nsr/nwx016
Publisher site
See Article on Publisher Site

### Abstract

Abstract Understanding the concentration and distribution of water in the Earth's mantle plays a substantial role in studying its chemical, physical and dynamic processes. After a decade of research, a comprehensive dataset of water content in upper-mantle samples has been built for eastern China, which is now the only place with water-content data from such diverse types of natural samples, and provides an integrated picture of the water content and its distribution in the upper mantle at a continental scale. The main findings include the following: (i) the temporal heterogeneity of the water content in the lithospheric mantle from early Cretaceous (∼120 Ma) to Cenozoic (<40 Ma) was tightly connected with the stability of the North China Craton (from its destruction to its consolidation); (ii) the heterogeneous water content in the Cenozoic lithospheric mantle beneath different blocks of eastern China was not only inherited from tectonic settings from which they came, but was also affected later by geological processes they experienced; (iii) the distinct water content between the lowermost crust and lithospheric mantle of eastern China and its induced rheological contrast at the base of the crust indicate that the continental crust–mantle boundary could behave either in a coupled or decoupled manner beneath different areas and/or at different stages; (iv) the alkali basalts of eastern China demonstrate a heterogeneous distribution of water content in the mantle; local and regional comparisons of the water content between the lithospheric mantle and basalts’ source indicate that the Cenozoic alkali basalts in eastern China were not sourced from the lithospheric mantle. Instead, the inferred high water contents in the mantle sources suggest that the Cenozoic eastern China basalts were likely sourced from the mantle transition zone (MTZ); and (v) both oceanic and continental crusts may carry a certain amount of water back into the deep mantle of eastern China by plate subduction. Such recycled crustal materials have not only created a local water-rich zone, but have also introduced crustal geochemical signatures into the mantle, both accounting for crustal geochemical imprints in the intra-plate magmatic rocks of eastern China. water, upper mantle and deep crust, continental stability, basalt genesis, eastern China INTRODUCTION Hydrogen that is structurally bound to other ions (mainly oxygen) in minerals is traditionally referred to as ‘water’ in the Earth science community and its concentration is calculated as H2O by weight. Its existence, even at trace levels (ppm level), can significantly influence certain important chemical and physical properties (e.g. seismic velocities, electrical and thermal conductivities, rheology, optical properties, melting temperature, pressure and phase relations and ion diffusion rate) of minerals and therefore rocks [1–13]. Consequently, water affects the chemical, physical and dynamic processes of rocks and their domains in the deep Earth, such as the relative movement of plates and the genesis and evolution of magmas [14–24]. Moreover, the amount of water in the continental lithospheric mantle is predicted to be closely related to its viscosity and stability [25,26]. Investigations on natural samples have demonstrated that an elevated water content can induce the destruction of cratons [27], whereas a reduced water content ensures craton longevity [22,28]. Therefore, understanding the concentration and distribution of water in the mantle plays a substantial role in explaining its formation and evolution. Furthermore, knowledge about the water content in the source of mantle-derived magmas may provide new constraints on the geochemical heterogeneities of the mantle and the involved processes [29,30]. The concentration and distribution of water in the deep lithosphere (lower crust and lithospheric mantle) and asthenospheric mantle can be determined by natural samples, including mafic granulites, peridotites, pyroxenites, eclogites and basalts, although information about the mantle transition zone (MTZ) and lower mantle is mostly constrained by high-temperature and -pressure experiments [31]. Recently, after a decade of research, a comprehensive dataset of the water content in peridotite and granulite xenoliths hosted by alkali basalts, terrain granulites, ultrahigh pressure metamorphic (UHPM) eclogites and Mesozoic-Cenozoic alkali basalts has been built for eastern China, which is now the only place with water-content data from such diverse types of natural samples (Fig. 1). This dataset provides an integrated picture of the water content and its distribution in the lower crust and upper mantle at a continental scale. In this review, we do not intend to compare our dataset from eastern China to those of other localities in the world because such comparisons have been covered well by Peslier [32] and Demouchy and Bolfan [33]. Instead, we will focus on the temporal and spatial distribution of water in the lower crust and upper mantle beneath eastern China and will summarize the origins of their heterogeneities and implications. Our primary purpose is to provide an example to address the water distribution and its role in the deep Earth at a continental scale. Figure 1. View largeDownload slide Sample locations in eastern China. (a) Peridotite and granulite locations; (b) basalt and eclogite locations. Figure 1. View largeDownload slide Sample locations in eastern China. (a) Peridotite and granulite locations; (b) basalt and eclogite locations. ANALYTICAL METHODS OF WATER CONTENT The main minerals in the Earth's upper mantle and lowermost crust include olivine (ol), clinopyroxene (cpx), orthopyroxene (opx), garnet (grt) and plagioclase (pl), and their high-pressure polymorphous. They are nominally anhydrous minerals (NAMs), meaning that there is no hydrogen in their stoichiometric chemical formulas. However, hydrogen can be incorporated into the structure of NAMs, usually as charge-compensating cation(s) often coupled with other cations (such as Al3+ or Ti4+). Hydrogen in such vacancies is normally bonded to one of the coordinating oxygens and forms a hydroxyl group [34], of which the content is expressed as H2O by weight ppm (hereafter referred to as ppm). Although the water content of minerals in the upper mantle and lowermost crust is generally less than several hundreds of ppm, they are likely the main water reservoir due to their predominant mass and volume proportions [32]. Fourier Transform Infrared Spectroscopy (FTIR) and Secondary Ion Mass Spectrometry (SIMS) are the most commonly used methods to measure the water content in NAMs [35]. Until now, all of the published water contents in mineral constituents of peridotites, granulites and eclogites from eastern China have been obtained using FTIR, with the exception of Aubaud et al. [36], who analysed several ol and pyroxene grains in Hannuoba peridotite xenoliths hosted in the Cenozoic basalts with SIMS. The SIMS data from Aubaud et al. [36] are comparable to the FTIR data of Yang et al. [37] from the same location. The water contents of the whole rocks, if reported, were estimated from the mineral water contents and their respective modal proportion through mass balance calculations. The water contents of basaltic magmas were calculated from the water content of cpx phenocrysts and the partition coefficient of water between cpx and basaltic melt. The details of the applied FTIR methodology can be found in Rossman [35], Kovács et al. [38], Xia et al. [27] and Demouchy and Bolfan [33]. The detailed run conditions and procedures have been described in the cited references. In short, the uncertainty of the measured water content of minerals in peridotites, granulites and eclogites was generally less than 20%, and that of the water content of basalts calculated from the water content of cpx phenocrysts was generally less than 40%. The factors influencing the uncertainty and the evaluation methods can be found in the cited references. WATER IN THE LITHOSPHERIC MANTLE OF EASTERN CHINA FTIR spectra and water contents The eastern Chinese continent consists of three main blocks from north to south: the Northeast China (NEC), the North China Craton (NCC) and the South China Block (SCB). Numerous small-volume basaltic volcanoes are distributed in these blocks and many contain abundant peridotite xenoliths, providing a good opportunity to investigate water distribution in the lithospheric mantle of eastern China. The water contents in the main constitute minerals (cpx, opx, ol and grt) in peridotite xenoliths from 28 basaltic localities, extending from Heilongjiang province in the north to Hainan province in the south (Fig. 1), have already been reported [37, 39–47]. Most of the studied peridotite xenoliths are spinel facies peridotites, except a few samples from the Nuomin volcano in the NEC and the Mingxi volcano in SCB, which are garnet facies peridotites [42,45]. FTIR investigations have shown that the garnets and most of the ol usually display no OH-related absorption bands, but there are also ol from several localities (Mingxi, Anyuan, Niutoushan and Qilin) in SCB that have prominent OH bands at 3575 cm−1, 3520 cm−1, 3340 cm−1 and 3230 cm−1. In contrast, except for the Nuomin peridotites, all of the opx and cpx show several OH absorption bands: 3595–3570 cm−1, 3525–3500 cm−1, 3415–3390 cm−1 and 3315–3300 cm−1 for opx, and 3635–3600 cm−1, 3550–3510 cm−1 and 3470–3445 cm−1 for cpx. The positions of these bands for ol, cpx and opx have been attributed to the vibration of the structural OH and are similar to those reported in peridotites worldwide [34,48–56]. It is particularly interesting that all of the ol, grt, cpx and opx grains in 13 peridotites from the Nuomin volcano in NEC have no any detectable OH band, demonstrating a very dry lithospheric mantle [45]. The water contents of minerals in the peridotites range from approximately 0 to 41 ppm for ol, 0 to 346 ppm for opx and 0 to 746 ppm for cpx. Except for some opx grains from the Panshishan and Tianchang basalts in the NCC [45,57], cpx and opx in peridotite xenoliths from other localities exhibit homogeneous water contents within individual grains, regardless of grain size. Moreover, the water content in opx usually displays positive correlations with the Al content [45,58]. Therefore, it is very likely that the cpx and opx retain the original water content that is typical for their mantle source. However, diffusional loss of hydrogen during peridotite ascent to the surface in the host basaltic magmas cannot be excluded for ol, considering near zero water in most of the ol grains. If Dcpx/ol = 10 is assumed, based on experimental results [20,21,36,39,59–63], the estimated water contents of the whole rocks using the calculated ol water content and the respective modal proportions of minerals range from 0 to 225 ppm, which cover the range of the Mid-ocean ridge basalt (MORB) source (50–200 ppm) [19,64,69] and are less than the ocean island basalt (OIB) source (300–1000 ppm) [68,70–74]. The water contents of the peridotites in eastern China vary among samples, even within a single locality. For example, the water contents range from 5 to 140 ppm, 5 to 355 ppm and 2 to 72 ppm in opx, cpx and whole rock, respectively, in peridotites from the Nushan volcano [37]. Furthermore, the water contents of the peridotites vary among different localities. For instance, the Nuomin peridotites in the NEC have no detectable water; the Hannuoba peridotites in the NCC have relatively low water content, ranging from 20 to 55 ppm, 50 to 150 ppm and 10 to 35 ppm for opx, cpx and whole rock, respectively; and the Jiande peridotites in the SCB contain much more water with 163 to 329 ppm, 388 to 589 ppm and 85 to 216 ppm for opx, cpx and whole rock, respectively [37,44]. Most of the ol grains in the peridotite xenoliths from eastern China have no detectable water; this is likely a result of the diffusional loss of hydrogen from ol during ascent, so the partition coefficient of water between ol and other minerals cannot be evaluated. The correlation of water content between cpx and opx is shown in Fig. 2 [36,40–47,58]. The ratios of H2O between cpx and opx fall in a range from 1.5 to 3.5 and can be regarded as partition coefficients (Dcpx/opx), since cpx and opx have preserved the original water content that is typical for their mantle source. The Dcpx/opx of the peridotites in eastern China has a similar range to those from natural peridotite xenoliths worldwide [22,34,50,52,53,55,56,75]. From the available experimental results [21,36,62,76,77], it appears that Dcpx/opx increases with increasing pressure (see the inserted plot in Fig. 2), so the variation in Dcpx/opx in peridotites worldwide may be a proxy for their depth of origin within the lithospheric mantle. If this is correct, it supports the common view of mantle petrologists and geochemists that the occasionally captured peridotite xenoliths by alkali magmas can well represent the entire lithospheric mantle, which is often questioned by geophysicists and geologists. Figure 2. View largeDownload slide Cpx water contents versus opx water contents in peridotite xenoliths of eastern China. The data are from Aubaud et al. [36], Yang et al. [58], Hao et al. [43–45,78], Li et al. [47], Wang et al. [46], Xia et al. [40,41] and Yu et al. [42]. The inserted plot shows the correlation between Dcpx/opx with pressure in experimental studies [21,36,62,76,77]. Figure 2. View largeDownload slide Cpx water contents versus opx water contents in peridotite xenoliths of eastern China. The data are from Aubaud et al. [36], Yang et al. [58], Hao et al. [43–45,78], Li et al. [47], Wang et al. [46], Xia et al. [40,41] and Yu et al. [42]. The inserted plot shows the correlation between Dcpx/opx with pressure in experimental studies [21,36,62,76,77]. Regional heterogeneity in the Cenozoic lithospheric mantle The regional distribution of the water content in the Cenozoic lithospheric mantle beneath all of eastern China, from the north part of the NEC (NNEC) to the SCB, has been discussed by Hao et al. [45] and is shown in Fig. 3. The Nuomin peridotite xenoliths are absolutely dry, indicating a dry lithospheric mantle beneath the NNEC. The lithospheric mantle of the south part of the NEC (SNEC) and the NCC is characterized by lower water contents than the MORB source, and the SNEC and the NCC share similar ranges and average values. The lithospheric mantle of the SCB has much higher water content than other regions in eastern China, falling in the range of the MORB source. By integrating the water content with other geochemical indices, Hao et al. [45] suggested that the regional variations in the water content of the lithospheric mantle beneath eastern China cannot be caused by distinct partial melting and mantle metasomatism events or the redox state. Instead, the lithospheric mantle beneath different regions may have distinct origins and have undergone distinct geodynamic processes. The NNEC lithospheric mantle is supposed to be from the Siberia craton, with hydrogen diffused out of the peridotite minerals when the Siberia craton had interacted with a super mantle plume. The relatively low water content of the NCC and the SNEC may have been caused by the reheating effect of an upwelling asthenospheric flow during the lithospheric thinning event in the Mesozoic. The higher water content of the SCB peridotites suggests that the large part of its deeper lithospheric mantle was accreted from the asthenosphere, and either the SCB did not undergo a significant lithospheric thinning event such as the NCC or the thinning mechanism was different. From the case of eastern China, it seems that water distribution in the lithospheric mantle may provide new constraints on the origins and geodynamic processes of continents. Figure 3. View largeDownload slide Comparison of the water contents of cpx, opx and whole rock of the peridotite xenoliths in different blocks from eastern China. The data sources are the same as in Fig. 2. Figure 3. View largeDownload slide Comparison of the water contents of cpx, opx and whole rock of the peridotite xenoliths in different blocks from eastern China. The data sources are the same as in Fig. 2. Temporal heterogeneity in the NCC lithospheric mantle Although the spatial distribution of water content and the controlling factors in the lithospheric mantle are of increasing interest to the Earth science community, less attention has been paid to the temporal variation of water content, which also matters greatly and may provide new insights into the evolution of the lithospheric mantle. Abundant mantle xenoliths entrained in the alkali basalts with a wide range of eruption ages (125 to < 40 Ma) across the NCC, especially the eastern part, provide a good chance to address that issue. A detailed investigation of the bulk water content of peridotite xenoliths from many localities indicates that the Cenozoic lithospheric mantle (<40 Ma) beneath the eastern part of the NCC was relatively dry, with an average of 25 ± 18 ppm (from 15 to 85 ppm), except for a few localities close to the deep-cutting Tan-Lu fault [40,45,58]. The extremely low bulk water content below 50 ppm compared to the samples from other continental lithospheric mantles worldwide that are typical for cratonic and off-cratonic peridotites (normally 40–180 ppm, with average values of 119 ± 54 ppm and 78 ± 45 ppm, respectively) and sub-oceanic mantle lithosphere values (>50 ppm) inferred from MORB and OIB (see the compiled dataset in Xia et al. [40]) highlights the uniqueness of the eastern part of the NCC and implies its links with the destruction of the NCC, as stated in the previous section. Li et al. [47] reported the water content of the peridotite xenoliths hosted by the late Mesozoic basalts (82–67 Ma) from eastern NCC, revealing two types of mantle domains with contrasting bulk water contents. The lithospheric mantle domain, represented by the Daxizhuang xenoliths at 74 Ma, was relatively ‘dry’, sharing similar characteristic with the Cenozoic lithospheric mantle [40,47,58]. Meanwhile, the peridotite xenoliths from Junan (67 Ma) and Qingdao (82 Ma) suggest a relatively high bulk water content for the lithospheric mantle, with an average of 130 ± 20 ppm, which is in the range of the MORB source (50–250 ppm). When mantle xenoliths are scarce, the bulk water content of the lithospheric mantle can also be obtained indirectly. Based on a study of water in the early crystallized cpx phenocrysts in basalts, Xia et al. [27] estimated that the bulk water content of the lithospheric mantle source of the Feixian high-magnesium basalts. The bulk water content was more than 1000 ppm, which is several times higher than that of the MORB source, demonstrating that the NCC lithospheric mantle was very water-enriched in the early Cretaceous (∼120 Ma), at the paroxysm of the destruction of the NCC. These combined snapshots envisage a temporal variation of water content for the eastern NCC lithospheric mantle (Fig. 4) [47]. At the peak of the destruction (∼120 Ma), the lithospheric mantle was hydrous; after its destruction was complete at the end of the Mesozoic, the water content of the lithospheric mantle was extremely low; during the period of destruction, the lithospheric mantle had intermediate water content [47]. Li et al. [47] further discussed the process of this cratonic destruction as follows: (i) at the peak time of destruction, the hydrated, water-rich lithospheric mantle offers a presumably low-viscosity state for its own destruction; (ii) during the destruction, the upwelling asthenosphere erodes the lowermost lithospheric mantle and continuously heats the overlying lithospheric mantle, which results in lithospheric thinning and dehydration of the relict lithospheric mantle. This goes on until the cessation of the destruction of the cratonic lithospheric mantle, which is, at least in part, due to its now lower water content and stiffer rheology, therefore supporting better resistance to further tectonic removal. The upwelling asthenosphere, however, cools and is transformed into a newly accreted lithospheric mantle with probably higher bulk water content than that of the dehydrated relict cratonic lithospheric mantle. Figure 4. View largeDownload slide The temporal variation of the bulk water content in the lithospheric mantle beneath the eastern NCC. The figure is modified from Fig. 11 in Li et al. [47] by adding the cpx water contents and the data sources are the same as Li et al. [47]. Figure 4. View largeDownload slide The temporal variation of the bulk water content in the lithospheric mantle beneath the eastern NCC. The figure is modified from Fig. 11 in Li et al. [47] by adding the cpx water contents and the data sources are the same as Li et al. [47]. These findings above seem to contrast with those of Liu and Xia [79], who reported the water content of six peridotite xenoliths entrained in the early Cretaceous (∼125 Ma) high-Mg diorite from Fushan in the western part of the NCC, which show much lower water content than their hydrated, more water-rich Mesozoic counterparts in the eastern part of the NCC. The water content in cpx and opx ranges from 216 to 404 ppm and 123 to 188 ppm, respectively, with an average of ∼40 ppm for whole rock, which is far less than >1000 ppm of the Feixian mantle source but is in the range of the classic craton [34,50,80,81]. The contrasting water contents could be interpreted as the hydration of the NCC lithospheric mantle by the westward subduction of the paleo-Pacific plate, which may not have yet reached the western part of the NCC by the early Cretaceous. WATER IN THE LOWER CONTINENTAL CRUST OF EASTERN CHINA Temporal variations in water content of the lower continental crust The lower continental crust, separating the shallow crust from the underlying lithospheric mantle, is of crucial importance in determining the tectonic evolution of continents and in buffering the exchange of materials between Earth's interior and exterior. Despite great efforts of water in mantle minerals, much less attention has been paid to the speciation and concentration levels of water in the lower crust and its bearing on physical properties. A survey by FTIR on xenolith and terrain granulites, typical of samples from the lower crust, from Hannuoba, Nushan and Daoxian, in eastern China, has shown that the main constitutive minerals, such as plagioclase (pl), cpx, opx and grt, commonly contain trace amounts of OH in their lattice structure, also including molecular H2O for some pl [58,82]. Thus, minerals in the lower crust have a similar ability to dissolve water as minerals in the upper mantle, although its concentration may be quite different (as shown below). The calculated water concentrations of lower crustal granulites range from approximately 200 to 2000 ppm for cpx, 60 to 1800 ppm for opx, 65 to 900 ppm for pl, 300 to 600 ppm for grt and 150 to 1100 ppm for the bulk rocks (Fig. 5) [37]. Recently, Németh et al. [83] measured the water contents of minerals in a suite of granulite xenoliths from the Pannonian Basin in east-central Europe; their data demonstrate the occurrence of approximately 0–271 ppm, 55–413 ppm and 0–684 ppm for cpx, opx and pl, respectively. These values are much smaller than those of other granulites from eastern China. Therefore, it appears that the distribution of water is highly heterogeneous in the lower crust, on both small (i.e. different samples from the same locality) and large (i.e. different localities) scales. This is similar to the observed chemical heterogeneities of many elements and isotopes in samples from the lower crust (e.g. Rudnick and Gao [84] and references therein). Such heterogeneities are presumably either inherited from the protoliths or they reflect the complicated history of crust–mantle interactions. Figure 5. View largeDownload slide Water contents of the lower crustal minerals in eastern China and their temporal contrast (after Yang et al. [37]). Figure 5. View largeDownload slide Water contents of the lower crustal minerals in eastern China and their temporal contrast (after Yang et al. [37]). An interesting result arising from the available data is that, by classifying the samples according to their formation ages, pre-Phanerozoic granulites record apparently higher water contents than Phanerozoic granulites (Fig. 5) [37]. One possible cause is that the early lower crust was more hydrous than the modern one, although it is not clear whether this is a local phenomenon or a general trend on a global scale [37]. For the granulite samples compiled in Fig. 5, the chemical composition is broadly similar for individual minerals, and so are the temperature and pressure conditions of equilibrium [37]. Therefore, a significant contrast in water content in these samples deserves some further discussion, even if Yang et al. [37] already provided some clues from a petrological perspective. Recently, Yang [85] experimentally demonstrated how, at a given pressure and temperature, the ability of pl in incorporating OH is more significant under very reduced conditions, e.g. at four or more log units below the QFM (quartz-fayalite-magnetite) buffer. Considering the thermodynamics of OH dissolution in silicate minerals [86] and the redox state of around QFM in the modern lower crust [87], the observed water contrast in Fig. 5 can be explained by a change of the prevailing redox state in the lower crust, in that the pre-Phanerozoic lower crust of eastern China was regionally more reduced. Alternatively, it is possible, considering the even lower water contents of granulite xenoliths from the younger Miocene Pannonian Basin in Central Europe [83], that the observed lower water content in the Phanerozoic Chinese xenolith may be related to repeated depletion events since the Precambrian. This appears to be a logical explanation, as the continental crust may have remained stable over geologic time, whereas the continental lithospheric mantle may have been removed during cratonic destruction. This means that the water budget in the continental lower crust may have decreased gradually over geologic time during repeated melt removals and heating episodes during extension and that it could only be rehydrated locally, as suggested by Németh et al. [83]. Vertical distribution of water content in the lower crust and lithospheric mantle The vertical distribution of water content in the shallow mantle has received increasing attention in recent years, and has been addressed in some studies [28]. However, the variation in water content from the lithospheric mantle to the overlying lower crust has not been well documented, even if it may be of equal importance for the stability and tectonics of the continents. Yang et al. [58] addressed this question with reported data of water contents from both granulite xenoliths and peridotite xenoliths from Hannuoba and Nushan within the NCC (eastern China). They found much higher water contents in both the mineral constituents and bulk rocks in granulites than in peridotites (Fig. 6) [58]. The contrast of water content may be related to petrological processes and histories of these rocks, such as partial melting, differentiation and metamorphism; however, thermodynamically, it may indicate that the incorporation mechanism of OH in lower crustal minerals is different from the mechanism in mantle minerals, e.g. due to different pressure, temperature and mineral chemistry (lower crustal pyroxenes are richer in Fe and Al than their mantle counterparts) [58]. Figure 6. View largeDownload slide Vertical distribution of water content in the lower crustal granulite xenoliths and mantle peridotite xenoliths. (a) The cpx and opx in the Hannuoba xenoliths; (b) the cpx and opx in the Nushan xenoliths; and (c) the bulk water content (after Yang et al. [58]). Figure 6. View largeDownload slide Vertical distribution of water content in the lower crustal granulite xenoliths and mantle peridotite xenoliths. (a) The cpx and opx in the Hannuoba xenoliths; (b) the cpx and opx in the Nushan xenoliths; and (c) the bulk water content (after Yang et al. [58]). The vertical variation in water content in the lower crust and sub-continental lithospheric mantle and their lateral differences, as shown in Fig. 6, allows us to explore the rheological properties of the deep lithosphere in eastern China. Considering the data on water and regional thermal (P–T) conditions, it appears that the lower crust is much stronger than the underlying shallow lithospheric mantle at Hannuoba but weaker at Nushan. This makes the lithosphere thermally and mechanically unstable at Hannuoba compared to Nushan. This indicates that, during the Cenozoic when the xenoliths were brought to the surface, the lithosphere may have undergone mechanical thinning at Hannuoba but thickening at Nushan [58]. This means that different lithospheric processes may have been operative in different tectonic areas beneath eastern China and that the variation in water content in the deep lithosphere is critical for understanding deep processes. WATER IN THE CENOZOIC ALKALI BASALTS OF EASTERN CHINA In eastern China, from Wudalianchi in Heilongjiang Province in north-eastern China to the Hainan Island in the south end of mainland China, the Cenozoic basalts are widely distributed. These basalts generally exhibit typical OIB-like trace element patterns and Sr–Nd isotopic compositions. The enriched components in the mantle sources of the Cenozoic basalts in eastern China are still not fully understood. The old lithospheric mantle, the recycled ancient lower continental crust and recycled oceanic crust in the asthenosphere were all suggested to be responsible for their formation [88–93]. Recently, the materials in the hydrated MTZ beneath eastern Asia were also proposed to be involved in the mantle sources of the basalts in north-eastern China [94,95]. It has been widely accepted that the water content of magma, and the corresponding H2O/Ce ratios, would be useful for identifying the mantle components [96–98]. The importance of water in our understanding of the genesis of alkali basalts in eastern China was also invoked as early as 2007 [99]. However, there has been minimal research trying to analyse the water contents of these basalts, except several attempts to argue that basalts from several locations may be rich in water that used thermodynamic calculations [95,100] or an indirect inference from the low δ18O values of mineral phenocrysts [101]. Only very recently were the water contents of the Cenozoic alkali basalts, from north-eastern to south-eastern China, constrained [29,30,102–105]. These results are based on a method that relies on the water contents of cpx phenocrysts and chemical composition-dependent water partitioning between cpx and basaltic melts, which was first suggested by Wade et al. [106] and considerably improved by Xia et al. [27] and Liu et al. [30]. Here, we summarize the distribution of the water contents in these alkali basalts in eastern China determined by this method and its inferences about the sources of the basalts, enriched components in the sources and implications about the role of the subducted Pacific slab on the genesis of the Cenozoic basalts in eastern Asia. Regional distribution of H2O and H2O/Ce It is worth mentioning that the uncertainty of this methodology may be up to ∼40% [30,41]. To recover the water content of the Cenozoic alkali basalts, cpx phenocrysts with Mg# (=100* Mg/(Mg+Fe), mol percent) >75 were used [29,30,104]. Although multiple processes, such as crystallization, crustal contamination, degassing and diffusion of hydrogen out of cpx during or after eruption, might affect the final calculated magma-water contents, reasonable correlations between the calculated water contents and incompatible element contents of bulk rocks (Fig. 7 [29,103] shows examples from the Shandong and Shuangliao basalts) indicate that none of these processes may have a significant impact [29,104]. The H2O/Ce ratios were determined by the calculated water contents and the measured Ce concentration of the bulk rocks, which are rather consistent with the average H2O/Ce values calculated by $$\frac{{C_w^{cpx}/D_w^{cpx/melt}}}{{C_{Ce}^{cpx}/D_{Ce}^{cpx/melt}}}$$ for a group of cpx phenocrysts, where C is the concentration of H2O and Ce, and D is the partition coefficient between cpx and basaltic melt [104]. Figure 7. View largeDownload slide The correlation between the calculated magma water contents and the bulk rock trace element concentrations. The data for the Shandong and Shuangliao basalts are from Liu et al. [104] and Chen et al. [29], respectively. Figure 7. View largeDownload slide The correlation between the calculated magma water contents and the bulk rock trace element concentrations. The data for the Shandong and Shuangliao basalts are from Liu et al. [104] and Chen et al. [29], respectively. The water contents and H2O/Ce ratios of the alkali basalts from different locations in eastern China are statistically shown in Fig. 8 [29,30,64–68,70,71,74,96–98,102–104,107–119], from which local and regional scale heterogeneities can be identified. The recovered magma-water contents varied from 0.6 to 3.9 wt.% in the Shandong basalts, from 0.9 to 3.1 wt.% in the Shuangliao basalts, from 1.1 to 2.7 wt.% in the Zhejiang basalts, from 0.2 to 3.8 wt.% in the Fujian basalts and from 1.6 to 4.2 wt.% in the Guangdong basalts. The higher water contents fall in the range of back-arc basin basalts (BABB) and island arc basalts (IAB), even considering the ∼40% uncertainty. By contrast, the water contents of the basalts in Taihang and Wulanhada in Central NCC and Xiaogulihe in the northern part of NE China are much lower (0.2–1.4 wt.%, 0.21–0.69 wt.% and ∼0.5 wt.%, respectively), which are close to the water contents of MORB and typical OIB. Note, the water contents found in the Wulanhada basalts do not support the conclusion that the basalts in this area are very hydrous [100,120], which was deduced from non-analytical approaches (the MELTs program modeling or indirect evidence from the involvement of hydrated lower oceanic crust, respectively). When the results from direct measurements are considered, it seems that the basalts in the region closer to the Pacific trench (Shuangliao, Shandong, Zhejiang, Fujian and Guangdong, referred to as the eastern area in Fig. 9) [120] exhibit higher water content in the magma than those further away from the trench (Xiaogulihe, Wulanhada and Taihang, referred to as the western area in Fig. 9). However, as an outlier, the water contents of the Chaihe-Aershan basalts in NE China span a similar range with those of the Shandong and Shuangliao basalts, although they have comparable distance to the Pacific trench as to the Wulanhada and Xiaogulihe basalts. Correspondingly, the H2O/Ce ratios of basalts exhibit a similar regional distribution (Fig. 8). When the maximum and mean values of the H2O/Ce ratios of the Shandong, Shuangliao, Zhejiang, Fujian, Guangdong and Chaihe-Aershan basalts are obviously higher than those of the most water-rich OIB (∼250) [98], those of the Taihang, Wulanhada and Xiaogulihe basalts are close to or significantly lower than that of the normal MORB (∼200–210) [122]. Figure 8. View largeDownload slide The water contents and H2O/Ce ratios of the Cenozoic alkali basalts in eastern China (after Chen et al. [121]). The data are presented from northern to southern localities from top to bottom. The water content data of the Cenozoic basalts in eastern China are from Chen et al. [29,102,103,121], Liu et al. [30,104] and Liu et al. [105]. The data source for IAB, BABB, OIB, MORB and E-MORB is from Asimow et al. [64], Danyushevsky et al. [107,108], Dixon and Clague [97], Dixon et al. [65,70,98], Dobson et al. [109], Hochstaedter et al. [110], Michael [66,96], Nichols et al. [71], Saal et al. [67], Simons et al. [68], Sisson and Layne [110], Sobolev and Chaussidon [69], Stolper and Newman [112] and Wallace [113,114]. The H2O/Ce ratio of PM is calculated with H2O from Dixon and Clague [97] and Ce from Sun and McDonough [115], and the H2O/Ce ratio of the EM1, EM2, HIMU, FOZO, DMM and IAB is from Cabral et al. [116], Dixon et al. [98], Kendrick et al. [117,118], Workman et al. [74] and Ruscitto et al. [119]. Figure 8. View largeDownload slide The water contents and H2O/Ce ratios of the Cenozoic alkali basalts in eastern China (after Chen et al. [121]). The data are presented from northern to southern localities from top to bottom. The water content data of the Cenozoic basalts in eastern China are from Chen et al. [29,102,103,121], Liu et al. [30,104] and Liu et al. [105]. The data source for IAB, BABB, OIB, MORB and E-MORB is from Asimow et al. [64], Danyushevsky et al. [107,108], Dixon and Clague [97], Dixon et al. [65,70,98], Dobson et al. [109], Hochstaedter et al. [110], Michael [66,96], Nichols et al. [71], Saal et al. [67], Simons et al. [68], Sisson and Layne [110], Sobolev and Chaussidon [69], Stolper and Newman [112] and Wallace [113,114]. The H2O/Ce ratio of PM is calculated with H2O from Dixon and Clague [97] and Ce from Sun and McDonough [115], and the H2O/Ce ratio of the EM1, EM2, HIMU, FOZO, DMM and IAB is from Cabral et al. [116], Dixon et al. [98], Kendrick et al. [117,118], Workman et al. [74] and Ruscitto et al. [119]. Figure 9. View largeDownload slide (a) Comparison of the water contents in the sources of the Cenozoic eastern China basalts with those in the peridotite xenoliths from the same regions. (b) The spatial distribution of the water content in the sources of the Cenozoic eastern China basalts. After Chen et al. [121]. Figure 9. View largeDownload slide (a) Comparison of the water contents in the sources of the Cenozoic eastern China basalts with those in the peridotite xenoliths from the same regions. (b) The spatial distribution of the water content in the sources of the Cenozoic eastern China basalts. After Chen et al. [121]. Where did water come from? Whether the Cenozoic basalts in eastern China are from the lithospheric mantle is still under fierce debate after more than three decades of intensive study [30,89,91,92,104,123–127]. Systematic investigations of the water contents of the alkali basalts and their hosted peridotite xenoliths that represent the lithospheric mantle may provide new insights. More precisely, our dataset on the water contents of eastern Chinese samples allow us to make both local and regional comparisons. This means that the water-content contrast between the basalt source and the lithospheric mantle for a specific area, and the regional water distribution in the lithospheric mantle and in the basalt source for entire eastern China could be explored. Chen et al. [121] have estimated the water contents in the sources of the Cenozoic eastern China basalts using the following: (i) the calculated water contents of the initial basaltic melts; (ii) the available partition coefficients of water between basalts and mantle rocks; and (iii) the estimated partial melting degrees that the basalt source experienced. The mantle source-water contents ranged between approximately 150 and 4700 ppm and most of them were higher than 500 ppm (Fig. 9b), regardless of the used partial melting model (batch melting or fractional melting) [121]. By comparing the estimated water contents in the basalt sources with those in the peridotite xenoliths [121], the flowing can be clearly determined: (i) the source-water contents for all the basalts are significantly higher (several to several hundred times) than those of the lithospheric mantle in the same region (Fig. 9a); (ii) except for the Chaihe-Arershan basalts in the north-west NEC, the water contents of the basalt sources appear to decrease from the east to the west (Fig. 9b), which is different from the water-content trend in the lithospheric mantle of eastern China decreasing from the south to the north (Fig. 3). Thus, the local and regional differences in the water contents in the lithospheric mantle and the basalt sources indicate that the Cenozoic basalts in eastern China may not have been derived from the lithospheric mantle. The upper mantle (except the lithospheric mantle containing amphibole or mica) can only accommodate 50–250 ppm of H2O [19,31,128], and the lower mantle contains less or similar amounts of water [128,129]. Only the MTZ can contain several hundred ppm to >10 000 ppm H2O [128,130]. Therefore, the estimated water contents (150–4700 ppm, and most of them higher than 500 ppm) in the sources of the Cenozoic eastern China basalts are likely from the MTZ. This proposition is compatible with geophysical investigations that have shown the stagnated oceanic slabs in the MTZ beneath eastern Asia [131,132], which are expected to provide water once they are disturbed. Enriched components in the mantle source The magma-water contents and corresponding H2O/Ce ratios provided new constraints on the origin of enriched components in the mantle sources of the Cenozoic eastern China basalts. As to the significance of the magma-water contents for identifying the source components, one typical example is the case of the Cenozoic alkali basalts in Shandong [104]. Both the recycled Pacific oceanic sediments and residual ancient lower continental crust (in the form of eclogite) that was produced by an earlier partial melting event have been suggested to be enriched components in the mantle source of the Cenozoic Shandong basalts [89,91,92,125]. Liu et al. [104] reported the water contents of the alkali basalts in one volcano (Fuyanshan volcano) in the same region. In addition to their exceptionally high water contents (see Regional distribution of H2O and H2O/Ce), the H2O/Ce ratios of these basalts were positively correlated with bulk-rock 87Sr/86Sr ratios and were negatively correlated with Nb/U ratios, respectively (Fig. 10) [104]. These trends demonstrate that the enriched components in the basalt sources should be enriched with water. Thus, the dry lower continental crust that experienced an earlier melting event before incorporation into the mantle sources could not be the appropriate candidate. Figure 10. View largeDownload slide The correlation of H2O/Ce with the bulk-rock 87Sr/86Sr and Nb/U ratios of the Fuyanshan basalts in Shandong. Modified from Liu et al. [104]. Figure 10. View largeDownload slide The correlation of H2O/Ce with the bulk-rock 87Sr/86Sr and Nb/U ratios of the Fuyanshan basalts in Shandong. Modified from Liu et al. [104]. The role of the Pacific subduction in the genesis of Cenozoic basalts in eastern China Many trace elemental, radiogenic and stable isotopic evidences have been found to support the mantle sources of basalts from ∼110 Ma to the late Cenozoic in eastern China that contain components from subducted oceanic slabs [91,101,133,134]. This hypothetical oceanic slab has usually been assigned to the subducted Pacific slab, which was stagnant at the MTZ beneath eastern Asia [131,135]. Recently, the changes in basalt geochemistry (such as Eu/Eu*, 87Sr/86Sr and δ18O values) with eruption ages were determined [92,126,136–138] and they were suggested to reflect the secular contributions of different portions of the subducted Pacific oceanic crust in the mantle sources. Very recently, similar temporal variations in magma-water contents for several Cenozoic basalts in eastern China were resolved. Chen et al. [29] reported the water contents of the Shuangliao basalts in NNEC, which erupted from 50 to 43 Ma. As shown in Fig. 11, the younger basalts exhibit higher H2O/Ce ratios along with higher Ba/Th and lower Ce/Pb ratios, which reflect a secular contribution of the subducted oceanic sediments in the mantle sources. The similar temporal variations of H2O/Ce and other trace elemental ratios were also observed for the basalts in the Chaihe-Aershan (∼1.27–0.25 Ma) volcanoes of north-east China and the Zhejiang (26–17 and 11 Ma) volcanoes of SCB [103,105]. Overall, all those recycled oceanic components in the mantle sources of the Cenozoic basalts were dynamically incorporated, which would be best linked to the on-going subduction of the Pacific slab. Figure 11. View largeDownload slide Correlations among H2O/Ce of the Cenozoic Shuangliao basalts and the bulk rock trace elemental ratios and radiogenic isotopic compositions. Gray squares represent the possible end members in the mantle source. Modified from Chen et al. [29]. Figure 11. View largeDownload slide Correlations among H2O/Ce of the Cenozoic Shuangliao basalts and the bulk rock trace elemental ratios and radiogenic isotopic compositions. Gray squares represent the possible end members in the mantle source. Modified from Chen et al. [29]. WATER IN THE DABIE-SULU UHPM ECLOGITES: MEANS OF CARRYING CRUSTAL WATER INTO THE DEEP Water content and distribution in the Dabie-Sulu UHPM eclogites It is well known that a great deal of aqueous fluid can be released from altered basalts and overlying sediments of the oceanic crust with increasing temperature and pressure during plate subduction. This process hydrates the overlying mantle wedge and usually generates syn-subduction arc magmatism on a large scale [139,140]. In contrast, aqueous fluids are less significant or absent during ultrahigh pressure metamorphism (UHPM) of the continental crust at mantle depths relative to that of the oceanic crust [141]. Because fluid is an important agent for element and isotope exchange, its presence or absence plays a critical role in the attainment and preservation of geochemical equilibrium between metamorphic minerals, and even the re-equilibration of geochronometric and geothermobarometric systems [142]. The Dabie-Sulu orogenic belt in east-central China is the biggest exposed UHPM belt in the world, providing a natural laboratory to investigate the transportation and recycling of water. In principle, the major constituent minerals of UHPM eclogites are NAMs such as garnet, omphacite, rutile and silica phases. They are all stable phases in the P–T range typical of continental subduction-zone metamorphism and likely represent the primary water reservoirs in subducted slabs. FTIR analyses of NAMs in the Dabie-Sulu UHPM eclogites demonstrate that they contain a certain amount of water as structural hydroxyl locked in vacancies, which can be up to ∼1915 ppm for garnet, ∼1850 ppm for omphacite, ∼9590 ppm for rutile and ∼440 ppm for coesite [143–148]. The similar quantities of water in both garnet and omphacite indicate that they play equivalent roles in transporting surface water to mantle depths during continental subduction (Fig. 12a) [34,143–157]. Similarly to oceanic subduction, such transported water by subducted continents would cause prominent hydration (at least locally) in the mantle which they interact with. Figure 12. View largeDownload slide Water contents in the Dabie-Sulu eclogites and comparisons with other UHPM eclogites and eclogite xenoliths. (a) water contents in garnet, omphacite and bulk rocks from the Dabie-Sulu orogenic belt; (b) water contents in garnets from the Dabie-Sulu eclogites and other UHPM eclogites and eclogite xenoliths; (c) water contents in omphacites from the Dabie-Sulu eclogites and other UHPM eclogites and eclogite xenoliths. The dataset for the Dabie-Sulu eclogites is from Zhang et al. [143,144], Xia et al. [145], Sheng et al. [146], Chen et al. [147], Zhao et al., [148] and Zhao and Zhang [149]; and the dataset for other UHPM eclogites and eclogite xenoliths is from Aines and Rossman [150], Bell and Rossman [34], Langer et al. [151], Snyder et al. [152], Matsyuk et al. [153], Ragozin et al. [154], Bell et al. [155], Katayama et al. [156] and Skogby et al. [157]. Figure 12. View largeDownload slide Water contents in the Dabie-Sulu eclogites and comparisons with other UHPM eclogites and eclogite xenoliths. (a) water contents in garnet, omphacite and bulk rocks from the Dabie-Sulu orogenic belt; (b) water contents in garnets from the Dabie-Sulu eclogites and other UHPM eclogites and eclogite xenoliths; (c) water contents in omphacites from the Dabie-Sulu eclogites and other UHPM eclogites and eclogite xenoliths. The dataset for the Dabie-Sulu eclogites is from Zhang et al. [143,144], Xia et al. [145], Sheng et al. [146], Chen et al. [147], Zhao et al., [148] and Zhao and Zhang [149]; and the dataset for other UHPM eclogites and eclogite xenoliths is from Aines and Rossman [150], Bell and Rossman [34], Langer et al. [151], Snyder et al. [152], Matsyuk et al. [153], Ragozin et al. [154], Bell et al. [155], Katayama et al. [156] and Skogby et al. [157]. Heterogeneous water contents in eclogite minerals were observed not only on the scale of the outcrop, but also among the different grains within the same hand-specimen. For instance, a >1000-ppm variation occurred for four to five garnet grains from ∼1-cm eclogite section [145]. Such small-scale heterogeneity clearly suggests very limited mobility of fluids during UHP metamorphism, and both subduction and exhumation processes of UHP rocks occurred over a short period of time [145]. Moreover, the decreased water content in the rims of some garnet and omphacite grains compared with the cores have been ascribed to hydroxyl exsolution upon the initial decompression exhumation of the UHPM eclogites. This process provides an efficient way to generate fluids during the early stage of exhumation and also promotes retrograde metamorphism [146]. The bulk-rock water contents are estimated to lie between 100 and 750 ppm [143–146]. In addition to structural OH, mineralogical studies by means of FTIR, transmission electron microscopy (TEM), backscattered electron (BSE) and laser Raman techniques indicate that NAMs contain significant amounts of water in the form of molecular H2O in fluid inclusions [145,158–161]. The fluid inclusions hosted by UHPM minerals may be either primary or secondary [162], thus providing an upper limit for the ability of transporting water into the mantle depth, while the structural hydroxyl in NAMs defines the lower limits. Comparison with upper-mantle eclogites It is worth noting that either consistency or inconsistency of water contents has been observed in UHPM eclogites relative to mantle eclogite xenoliths hosted by kimberlites. For garnets, the water contents of UHP metamorphic rocks are significantly higher than in eclogite xenoliths (Fig. 12b). The prolonged residence in mantle conditions of eclogite xenoliths has the potential to result in a significant loss of water to the surrounding peridotitic mantle due to the usually much lower water contents in upper-mantle peridotites. Consequently, this can explain the lower water contents in garnets from eclogitic xenoliths [145]. In contrast, due to the presumably much shorter duration of tectonic exhumation, the UHPM eclogites may have a better capacity to preserve the water-rich character of their protoliths [145,146,163]. However, the water contents in omphacites from UHPM eclogites are comparable to those in eclogite xenoliths (Fig. 12c). This may reflect both the pressure and composition control on the incorporation of structural hydroxyl in crystal defects such as the Ca-Eskola components in M2-site vacancies [157,164]. Moreover, jadeite with water contents ranging from 100 to 1950 ppm and coherent variations in Na and Ca contents, as well as M2-site vacancies, have been reported for the Dabie UHPM jadeite-bearing quartzites [165]. These characteristics suggest that bulk mineral composition plays an important role in the incorporation of hydrogen [165]. Taken as a whole, the UHPM eclogites can contain higher water contents (>200 ppm) than MORB-sourced mantle peridotites that range from 50 to 200 ppm [166]. Therefore, the breakdown of such a water-rich subducted continental slab would have the potential to contribute to the ‘hidden’ water-rich region in mantle depths as the OIB source. Water in the form of structural hydroxyl is usually immobile in both hydrous and anhydrous minerals in UHPM conditions. The breakdown of the hydrous minerals in UHPM slices provides a dominant source for aqueous fluid during continental subduction-zone metamorphism [167,168]. Furthermore, numerous experimental studies have demonstrated that the solubility of structural hydroxyl in NAMs increases with pressure [156,164,169–175]. This, in turn, implies that the structural hydroxyl can be released from NAMs during exhumation. Therefore, significant amounts of aqueous fluids are available from the exsolution of structural hydroxyl and molecular water from NAMs in the initial stage of exhumation. The aqueous fluid then reacts with its host minerals resulting in their dissolution and recrystallization. Consequently, this metamorphic dehydration during subduction/exhumation would probably provide a sufficient amount of aqueous fluid for not only high-pressure eclogite-facies quartz veining, but also amphibolite-facies retrogression. SUMMARIZED CONCLUSIONS Although the main minerals in the lithospheric mantle of eastern China are nominally anhydrous, they usually contain a certain amount of water (up to several hundreds of ppm H2O) as hydroxyl groups in their structural defects. The temporal variation of the water content in the lithospheric mantle from the early Cretaceous (∼120 Ma) to Cenozoic (<40 Ma) appears tightly connected to the stability of the NCC (from its destruction to its consolidation), thus providing a clear case to show that the rheological changes induced by water in the lithospheric mantle affect the stability of continents. The heterogeneous water content in the Cenozoic lithospheric mantle beneath different blocks of eastern China was not only inherited from tectonic settings from which they came before having been amalgamated into a single continent, but was also affected by later geological processes that they experienced. The distinct water content between the lowermost crust and lithospheric mantle of eastern China and its induced rheological contrast at the base of the crust indicate that the continental crust–mantle boundary could behave either in a coupled or decoupled style beneath different areas and/or at different stages. The water contrast between the Precambrian and Phanerozoic lower crust of eastern China suggests a temporal evolution of the water content in the Earth's crust, different formation mechanisms of the continental lower crust or both. It is likely that the continental lower crust may have become gradually depleted with respect to its water content over geologic time, contributing to the stronger rheology and, therefore, preservation of the continental crust. The alkali basalts of eastern China demonstrate a heterogeneous distribution of the water content in the upper mantle and MTZ. Pacific plate subduction is likely the main process to introduce heterogeneities in the water content and other geochemical characteristics of the intra-plate magmas. Local and regional comparisons of the water content between the lithospheric mantle and basalt source demonstrate that the Cenozoic alkali basalts in eastern China were not sourced from the lithospheric mantle. Instead, the inferred high water contents in the mantle sources indicate that the Cenozoic eastern China basalts were likely sourced from the MTZ. The UHPM eclogites of the Dabie-Sulu orogeny can host up to thousands of ppm of water in nominally anhydrous garnets and omphacites, demonstrating that the ‘dry’ continental crust may carry a certain amount of water back into the deep mantle by subduction. Such recycled continental crustal materials not only created a local water-rich zone, but also delivered continental geochemical signatures into the mantle, both accounting for the continental geochemical imprints in some OIBs and other intra-plate basalts. FUTURE ISSUES More data from other areas are necessary to examine certain conclusions drawn from the samples of eastern China, such as the higher water content in the lower crust than in the underlying lithospheric mantle and the more hydrous Precambrian lower crust than its Phanerozoic counterpart. In addition, the spatial and temporal distribution of the water content in the lithospheric mantle and its effect on continental evolution also require further exploration by collecting more data from other parts of the world. Pyroxenites are an important rock constituent in the upper mantle, and are likely the source lithology of some intra-plate basalts; in particular, local enrichment of Fe- and H-bearing pyroxenites may lead to regionally electrical anomalies, affecting the structure and some key properties of the upper mantle [176]. However, less attention has been paid to water in pyroxenites, and so far only Bizmis and Peslier [177] have reported water in six pyroxenite xenoliths with similar genesis (crystallized from basaltic magmas) from Hawaiian basalts. Due to the prominent role of water in the temperature, pressure and degree of melting and composition of the yielded melts, it is necessary to systematically study water in pyroxenites in different tectonic settings. Felsic rocks are the main constituents in UHPM orogens and the dominant assemblages in the shallower crust and the content and partitioning of water in minerals of felsic rocks should be investigated for a more comprehensive understanding of the recycling of water during the subduction and exhumation of continental plates and the storage of water in the crust and its exchange between different reservoirs. The magma-water contents provide new constraints on some important issues related to the production of intra-plate continental basalts, such as the nature of mantle sources (lithosphere vs. asthenosphere) and origin of enriched components. Available research relies on the established cpx phenocryst-based approach. The uncertainty of this method is relatively large and this method cannot be applied to the basalts free of cpx phenocrysts. Thus, in the future, a new methodology with better accuracy and wide applicability should be developed. As we demonstrated, the partition coefficient of water between opx and cpx in peridotites shows a range of values. In the future, these values should be more precisely explored with an explanation for scattering and whether there is any relationship with geochemical and physical variables. Acknowledgments We thank Dr Zhong-He Zhou for inviting us to write this paper. Three anonymous referees are acknowledged for thorough reviews and providing constructive suggestions. FUNDING This work was supported by the National Natural Science Foundation of China (41225005). Conflict of interest statement. None declared. REFERENCES 1. Mackwell SJ , Kohlstedt DL , Paterson MS . The role of water in the deformation of olivine single-crystals . J Geophys Res 1985 ; 90 : 1319 – 33 . Google Scholar CrossRef Search ADS 2. Karato S , Paterson MS , Fitz-Gerald JD . Rheology of synthetic olivine aggregates: influence of grains size and water . J Geophys Res 1986 ; 91 : 8151 – 76 . Google Scholar CrossRef Search ADS 3. Karato S . The role of hydrogen in the electrical conductivity of the upper mantle . Nature 1990 ; 347 : 272 – 3 . Google Scholar CrossRef Search ADS 4. Graham CM , Elphick SC . Some experimental constraints on the role of hydrogen in oxygen and hydrogen diffusion and Al-Si interdiffusion in silicates . In: Ganguly J (ed.) . Diffusion, atomic ordering, and mass transport selected topics in geochemistry. Adv Phys Geochem 1991 ; 8 : 248 – 85 . Google Scholar CrossRef Search ADS 5. Meade C , Jeanloz R . Deep-focus earthquakes and recycling of water into the Earth's mantle . Science 1991 ; 252 : 68 – 72 . Google Scholar CrossRef Search ADS PubMed 6. Inoue T . Effect of water on melting phase relations and melt compositions in the system Mg2SiO4-MgSiO3-H2O up to 15 GPa . Phys Earth Planet Inter 1994 ; 85 : 237 – 63 . Google Scholar CrossRef Search ADS 7. Hirose K . Melting experiments on lherzolite KLB-1 under hydrous conditions and generation of high-magnesium andesitic melts . Geology 1997 ; 25 : 42 – 4 . Google Scholar CrossRef Search ADS 8. Mei S , Kohlstedt DL . Influence of water on plastic deformation of olivine aggregates. I: Diffusion creep regime . J Geophys Res 2000 ; 105 : 21457 – 69 . Google Scholar CrossRef Search ADS 9. Jung H , Karato S . Water-induced fabric transitions in olivine . Science 2001 ; 293 : 1460 – 3 . Google Scholar CrossRef Search ADS PubMed 10. Hofmeister AM . Enhancement of radiative transfer in the upper mantle by OH- in minerals . Phys Earth Planet Inter 2004 ; 146 : 483 – 95 . Google Scholar CrossRef Search ADS 11. Wang Z , Hiraga T , Kohlstedt DL . Effect of H+ on Fe-Mg interdiffusion in olivine, (Fe, Mg)2SiO4 . Appl Phys Lett 2004 ; 85 : 209 – 11 . Google Scholar CrossRef Search ADS 12. Hier-Majumder S , Mei S , Kohlstedt DL . Water weakening of clinopyroxeneite in diffusion creep . J Geophy Res 2005 ; 110 : B07406 . Google Scholar CrossRef Search ADS 13. Karato SI . Rheology of the deep upper mantle and its implications for the preservation of the continental roots: a review . Tectonophysics 2010 ; 481 : 82 – 98 . Google Scholar CrossRef Search ADS 14. Gaetani GA , Grove TL , Bryan WB . 1993 . The influence of water on the petrogenesis of subduction related igneous rocks . Nature 365 : 332 – 4 . Google Scholar CrossRef Search ADS 15. Hirose K , Kawamoto T . Hydrous partial melting of lherzolite at 1 GPa: The effect on H2O on the genesis of basaltic magmas . Earth Planet Sci Lett 1995 ; 133 : 463 – 73 . Google Scholar CrossRef Search ADS 16. Hirth G , Kohlstedt DL . Water in the oceanic upper mantle: Implications for rheology, melt extraction and the evolution of the lithosphere . Earth Planet Sci Lett 1996 ; 144 : 93 – 108 . Google Scholar CrossRef Search ADS 17. Asimow PD , Langmuir CH . 2003 . The important of water to oceanic mantle melting regimes . Nature 421 : 815 – 20 . Google Scholar CrossRef Search ADS PubMed 18. Dixon JE , Dixon TH , Bell DR et al. Lateral variation in upper mantle viscosity: role of water . Earth Planet Sci Lett 2004 ; 222 : 451 – 67 . Google Scholar CrossRef Search ADS 19. Hirschmann MM , Aubaud C , Withers AC . Storage capacity of H2O in nominally anhydrous minerals in the upper mantle . Earth Planet Sci Lett 2005 ; 236 : 167 – 81 . Google Scholar CrossRef Search ADS 20. Hirschmann MM , Tenner T , Aubaud C et al. Dehydration melting of nominally anhydrous mantle: the primacy of partitioning . Phys Earth Planet Inter 2009 ; 176 : 54 – 68 . Google Scholar CrossRef Search ADS 21. Hauri EH , Gaetani GA , Green TH . Partitioning of water during melting of the Earth's upper mantle at H2O-undersaturated conditions . Earth Planet Sci Lett 2006 ; 248 : 715 – 34 . Google Scholar CrossRef Search ADS 22. Li ZX , Lee CT , Peslier AH et al. Water contents in mantle xenoliths from the Colorado Plateau and vicinity: Implications for the mantle rheology and hydration-induced thinning of continental lithosphere . J Geophys Res 2008 ; 13 , B0921010.1029/2007jb005540 . 23. Grove TL , Till CB , Krawczynski MJ . The role of H2O in subduction zone magmatism . Annu Rev Earth Planet Sci 2012 ; 40 : 413 – 39 . Google Scholar CrossRef Search ADS 24. Green DH . Experimental petrology of peridotites, including effects of water and carbon on melting in the Earth's upper mantle . Phys Chem Minerals 2015 ; 42 : 95 – 122 . Google Scholar CrossRef Search ADS 25. Jordan TH . Composition and development of continental tectosphere . Nature 1978 ; 274 : 544 – 8 . Google Scholar CrossRef Search ADS 26. Pollack HN . Cratonization and thermal evolution of the mantle . Earth Planet Sci Lett 1986 ; 80 : 175 – 82 . Google Scholar CrossRef Search ADS 27. Xia QK , Liu J , Liu SC et al. High water content in Mesozoic primitive basalts of the North China Craton and implications on the destruction of cratonic mantle lithosphere . Earth Planet Sc Lett 2013 ; 361 : 85 – 97 . Google Scholar CrossRef Search ADS 28. Peslier AH , Woodland AB , Bell DR et al. Olivine water contents in the continental lithosphere and the longevity of cratons . Nature 2010 ; 467 : 78 – U108 . Google Scholar CrossRef Search ADS PubMed 29. Chen H , Xia QK , Ingrin J et al. Changing recycled oceanic components in the mantle source of the Shuangliao Cenozoic basalts, NE China: New constraints from water content . Tectonophysics . 2015 ; 650 : 113 – 23 . Google Scholar CrossRef Search ADS 30. Liu J , Xia QK , Deloule E et al. Recycled oceanic crust and marine sediment in the source of alkali basalts in Shandong, eastern China: evidence from magma water content and oxygen isotopes . J Geophys Res 2015 ; 120 : 8281 – 303 . Google Scholar CrossRef Search ADS 31. Ohtani E . Hydrous minerals and the storage of water in the deep mantle . Chem Geol 2015 ; 418 : 6 – 15 . Google Scholar CrossRef Search ADS 32. Peslier AH . A review of water contents of nominally anhydrous minerals in the mantles of Earth, Mars and the Moon . J Vol Geothem Res 2010 ; 197 : 239 – 58 . Google Scholar CrossRef Search ADS 33. Demouchy S , Bolfan-Casanova N . Distribution and transport of hydrogen in the lithospheric mantle: A Review . Lithos 2016 ; 240-243 : 402 – 25 . Google Scholar CrossRef Search ADS 34. Bell DR , Rossman GR . Water in Earth's mantle: the role of nominally anhydrous minerals . Science 1992 ; 255 : 1391 – 7 . Google Scholar CrossRef Search ADS PubMed 35. Rossman GR . Analytical methods for measuring water in nominally anhydrous minerals . Rev Mineral Geochem 2006 ; 62 : 1 – 28 . Google Scholar CrossRef Search ADS 36. Aubaud C , Hauri EH , Hirschmann MM . Hydrogen partition coefficients between nominally anhydrous minerals and basaltic melts . Geophys Res Lett 2004 ; 31 : 2 – 5 . Google Scholar CrossRef Search ADS 37. Yang XZ , Deloule E , Xia QK et al. Water contrast between Precambrian and Phanerozoic continental lower crust in eastern China . J Geophys Res 2008 ; 113 : B08207 . Google Scholar CrossRef Search ADS 38. Kovács I , Hermann J , O’Neill HSC et al. Quantitative absorbance spectroscopy with unpolarized light: Part II. Experimental evaluation and development of a protocol for quantitative analysis of mineral IR spectra . Am Mineral 2008 ; 93 : 765 – 78 . Google Scholar CrossRef Search ADS 39. Aubaud C , Withers AC , Hirschmann MM et al. Intercalibration of FTIR and SIMS for hydrogen measurements in glasses and nominally anhydrous minerals . Am Mineral 2007 ; 92 : 811 – 28 . Google Scholar CrossRef Search ADS 40. Xia QK , Hao YT , Li P et al. Low water content of the Cenozoic lithospheric mantle beneath the eastern part of the North China Craton . J Geophy Res 2010 ; 115 : B07207 . Google Scholar CrossRef Search ADS 41. Xia QK , Hao YT , Liu SC et al. Water contents of the Cenozoic lithospheric mantle beneath the western part of the North China Craton: Peridotite xenolith constraints . Gond Res 2013 ; 23 : 108 – 18 . Google Scholar CrossRef Search ADS 42. Yu Y , Xu XS , Griffin WL et al. H2O contents and their modification in the Cenozoic subcontinental lithospheric mantle beneath the Cathaysia block, SE China . Lithos 2011 ; 126 : 182 – 97 . Google Scholar CrossRef Search ADS 43. Hao YT , Xia QK , Liu SC et al. Recognizing juvenile and relict lithospheric mantle beneath the North China Craton: combined analysis of H2O major and trace elements and Sr-Nd isotope compositions of clinopyroxenes . Lithos 2012 ; 149 : 136 – 45 . Google Scholar CrossRef Search ADS 44. Hao YT , Xia QK , Li QW et al. Partial melting control of water contents in the Cenozoic lithospheric mantle of the Cathaysia block of South China . Chem Geol 2014 ; 380 : 7 – 19 . Google Scholar CrossRef Search ADS 45. Hao YT , Xia QK , Jia ZB et al. Regional heterogeneity in the water content of the Cenozoic lithospheric mantle of Eastern China . J Geophys Res 2016 ; 121 : 1 – 21 . 46. Wang Q , Bagdassarov N , Xia QK et al. Water contents and electrical conductivity of peridotite xenoliths from the North China Craton: implications for water distribution in the upper mantle . Lithos 2014 ; 189 : 105 – 26 . Google Scholar CrossRef Search ADS 47. Li P , Xia QK , Deloule E et al. Temporal variation of H2O content in the lithospheric mantle beneath the eastern North China Craton: implications for the destruction of cratons . Gond Res 2015 ; 28 : 276 – 87 . Google Scholar CrossRef Search ADS 48. Demouchy S , Ishikawa A , Tommasi A et al. Characterization of hydration in the mantle lithosphere: peridotite xenoliths from the Ontong Java Plateau as an example . Lithos 2015 ; 212-215 : 189 – 201 . Google Scholar CrossRef Search ADS 49. Denis CMM , Alard O , Demouchy S . Water content and hydrogen behaviour during metasomatism in the uppermost mantle beneath Ray Pic volcano (Massif Central, France) . Lithos 2015 ; 236–7 : 256 – 74 . Google Scholar CrossRef Search ADS 50. Grant K , Ingrin J , Lorand JP et al. Water partitioning between mantle minerals from peridotite xenoliths . Contrib Mineral Petrol 2007 ; 154 : 15 – 34 . Google Scholar CrossRef Search ADS 51. Ingrin J , Skogby H . Hydrogen in nominally anhydrous upper-mantle minerals: concentration levels and implications . Eur J Mineral 2000 ; 12 : 543 – 70 . Google Scholar CrossRef Search ADS 52. Peslier AH , Luhr JF , Post J . Low water contents in pyroxenes from spinel-peridotites of the oxidized, sub-arc mantle wedge . Earth Planet Sci Lett 2002 ; 201 : 69 – 86 . Google Scholar CrossRef Search ADS 53. Peslier AH , Woodland AB , Bell DR et al. Metasomatic control of water contents in the Kaapvaal cratonic mantle . Geochim Cosmochim Acta 2012 ; 97 : 213 – 46 . Google Scholar CrossRef Search ADS 54. Skogby H , Bell DR , Rossman GR . Hydroxide in pyroxene: variation in the natural enviroment . Am Mineral 1990 ; 75 : 764 – 74 . 55. Sundvall R , Stalder R . Water in upper mantle pyroxene megacrysts and xenocrysts: a survey study . Am Mineral 2011 ; 96 : 1215 – 27 . Google Scholar CrossRef Search ADS 56. Warren J , Hauri E . Pyroxenes as tracers of mantle water variations . J Geophy Res 2014 ; 119 : 1851 – 81 . Google Scholar CrossRef Search ADS 57. Tian ZZ , Liu J , Xia QK et al. Water concentration profiles in natural mantle orthopyroxenes: a geochronometer for long annealing of xenoliths within magma . Geology 2017 ; 45 : 87 – 90 . Google Scholar CrossRef Search ADS 58. Yang XZ , Xia QK , Deloule E et al. Water in minerals of continental lithospheric mantle and overlying lower crust: a comparative study of peridotite and granulite xenoliths from the North China Craton . Chem Geol 2008 ; 256 : 33 – 45 . Google Scholar CrossRef Search ADS 59. Koga K , Hauri E , Hirschmann M et al. Hydrogen concentration analyses using SIMS and FTIR: comparison and calibration for nominally anhydrous minerals . Geochem Geophys Geosyst 2003 ; 4 , 10.1029/2002GC000378 . 60. Withers AC , Hirschmann MM . H2O storage capacity of MgSiO3 clinoenstatite at 8-13 GPa, 1,100-1,400°C . Contrib Mineral Petrol 2007 ; 154 : 663 – 74 . Google Scholar CrossRef Search ADS 61. Withers AC , Hirschmann MM . Influence of temperature, composition, silica activity and oxygen fugacity on the H2O storage capacity of olivine at 8 GPa . Contrib Mineral Petrol 2008 ; 156 : 595 – 605 . Google Scholar CrossRef Search ADS 62. Tenner TJ , Hirschmann MM , Withers AC et al. Hydrogen partitioning between nominally anhydrous upper mantle minerals and melt between 3 and 5 GPa and applications to hydrous peridotite partial melting . Chem Geol 2009 ; 262 : 42 – 56 . Google Scholar CrossRef Search ADS 63. Adam J , Turner M , Hauri EH et al. Crystal/melt partitioning of water and other volatiles during the near-solidus melting of mantle peridotite: comparisons with non-volatile incompatible elements and implications for the generation of intraplate magmatism . Am Mineral 2016 ; 101 : 876 – 88 . Google Scholar CrossRef Search ADS 64. Asimow PD , Dixon JE , Langmuir CH . A hydrous melting and fractionation model for mid-ocean ridge basalts: application to the Mid-Atlantic Ridge near the Azores . Geochem Geophy Geosyst 2004 ; 5 , 10.1029/2003GC000568 . 65. Dixon JE , Stolper E , Delaney JR . Infrared spectroscopic measurements of CO2 and H2O in Juan de Fuca Ridge basaltic glasses . Earth Planet Sci Lett 1988 ; 90 : 87 – 104 . Google Scholar CrossRef Search ADS 66. Michael PJ . The concentration, behavior and storage of H2O in the suboceanic upper mantle: implications for mantle metasomatism . Geochim Cosmochim Acta 1988 ; 52 : 555 – 66 . Google Scholar CrossRef Search ADS 67. Saal AE , Hauri EH , Langmuir CH et al. Vapour undersaturation in primitive mid-ocean-ridge basalt and the volatile content of Earth's upper mantle . Nature 2002 ; 419 : 451 – 5 . Google Scholar CrossRef Search ADS PubMed 68. Simons K . Volatiles in basaltic glasses from the Easter-Salas y Gomez Seamount Chain and Easter Microplate: implications for geochemical cycling of volatile elements . Geochem Geophys Geosyst 2002 ; 3 , 10.1029/2001GC000173 . 69. Sobolev AV , Chaussidon M . H2O concentrations in primary melts from supra-subduction zones and mid-ocean ridges: implications for H2O storage and recycling in the mantle . Earth Planet Sci Lett 1996 ; 137 : 45 – 55 . Google Scholar CrossRef Search ADS 70. Dixon JE , Clague DA , Wallace P et al. Volatiles in alkalic basalts from the North Arch volcanic field, Hawaii: extensive degassing of deep submarine-erupted alkalic series lavas . J Petrol 1997 ; 38 : 911 – 39 . Google Scholar CrossRef Search ADS 71. Nichols ARL , Carroll MR , Höskuldsson Á . Is the Iceland hot spot also wet? Evidence from the water contents of undegassed submarine and subglacial pillow basalts . Earth Planet Sci Lett 2002 ; 202 : 77 – 87 . Google Scholar CrossRef Search ADS 72. Seaman C , Sherman SB , Garcia MO et al. Volatiles in glasses from the HSDP2 drill core . Geochem Geophys Geosystems 2004 ; 5 , 10.1029/2003GC000596 . 73. Wallace PJ . Volatiles in submarine basaltic glasses from the Northern Kerguelen Plateau (ODP Site 1140): implications for source region compositions, magmatic processes, and plateau subsidence . J Petrol 2002 ; 43 : 1311 – 26 . Google Scholar CrossRef Search ADS 74. Workman RK , Hauri E , Hart SR et al. Volatile and trace elements in basaltic glasses from Samoa: implications for water distribution in the mantle . Earth Planet Sci Lett 2006 ; 241 : 932 – 51 . Google Scholar CrossRef Search ADS 75. Falus G , Tommasi A , Ingrin J et al. Deformation and seismic anisotropy of the lithospheric mantle in the southeastern Carpathians inferred from the study of mantle xenoliths . Earth Planet Sci Lett 2008 ; 272 : 50 – 64 . Google Scholar CrossRef Search ADS 76. Kovács I , Green DH , Rosenthal A et al. An experimental study of water in nominally anhydrous minerals in the upper mantle near the water-saturated solidus . J Petrol 2012 ; 53 : 2067 – 93 . Google Scholar CrossRef Search ADS 77. Novella D , Frost DJ . The composition of hydrous partial melts of garnet peridotite at 6 GPa: implications for the origin of Group II Kimberlites . J Petrol 2014 ; 55 : 2097 – 124 . Google Scholar CrossRef Search ADS 78. Hao YT , Xia QK , Tian ZZ et al. Mantle metasomatism did not modify the water content of the peridotite xenoliths from the Tianchang basalts of eastern China . Lithos 2016 ; 260 : 315 – 27 . Google Scholar CrossRef Search ADS 79. Liu SC , Xia QK . Water content in the early Cretaceous lithospheric mantle beneath the south-central Taihang Mountains: implications for the destruction of the North China Craton . Chin Sci Bull 2014 ; 59 : 1362 – 5 . Google Scholar CrossRef Search ADS 80. Peslier AH , Woodland AB , Wolff JA . Fast kimberlite ascent rates estimated from hydrogen diffusion profiles in xenolithic mantle olivines from southern Africa . Geochim Cosmochim Acta 2008 ; 72 : 2711 – 22 . Google Scholar CrossRef Search ADS 81. Baptiste V , Tommasi A , Demouchy S . Deformation and hydration of the lithospheric mantle beneath the Kaapvaal craton, South Africa . Lithos 2012 ; 149 : 31 – 50 . Google Scholar CrossRef Search ADS 82. Xia QK , Yang X , Deloule E et al. Water in the lower crustal granulite xenoliths from Nushan, eastern China . J Geophys Res 2006 ; B11202 , 10.1029/2006JB004296 . 83. Németh B , Török K , Kovács I et al. Melting, fluid migration and fluid-rock interactions in the lower crust beneath the Bakony-Balaton Highland volcanic field: a silicate melt and fluid inclusion study . Miner Petrol 2015 ; 109 : 217 – 34 . Google Scholar CrossRef Search ADS 84. Rudnick RL , Gao S . Composition of the continental crust . In: Rudnick RL (ed.) . Treatise on Geochemistry: The Crust . Oxford: Elsevier-Pergamon , 2003 , 1 – 64 . 85. Yang XZ . An experimental study of H solubility in feldspars: effect of composition, oxygen fugacity, temperature and pressure and implications for crustal processes . Geochim Cosmochim Acta 2012 ; 97 : 46 – 57 . Google Scholar CrossRef Search ADS 86. Keppler H , Bolfan-Casanova N . Thermodynamics of water solubility and partitioning . Rev Mineral Geochem 2006 ; 62 : 193 – 230 . Google Scholar CrossRef Search ADS 87. Yang XZ , Gaillard F , Scaillet B . A relatively reduced Hadean continental crust and implications for the early atmosphere and crustal rheology . Earth Planet Sci Lett 2014 ; 393 : 210 – 9 . Google Scholar CrossRef Search ADS 88. Tang YJ , Zhang HF , Ying JF . Asthenosphere-litho spheric mantle interaction in an extensional regime: implication from the geochemistry of Cenozoic basalts from Taihang Mountains, North China Craton . Chem Geol 2006 ; 233 : 309 – 27 . Google Scholar CrossRef Search ADS 89. Zeng G , Chen LH , Hofmann AW et al. Crust recycling in the sources of two parallel volcanic chains in Shandong, North China . Earth Planet Sci Lett 2011 ; 302 : 359 – 68 . Google Scholar CrossRef Search ADS 90. Chen LH , Zeng G , Jiang SY et al. Sources of Anfengshan basalts: subducted lower crust in the Sulu UHP belt, China . Earth Planet Sc Lett . 2009 ; 286 : 426 – 35 . Google Scholar CrossRef Search ADS 91. Zhang JJ , Zheng YF , Zhao ZF . Geochemical evidence for interaction between oceanic crust and lithospheric mantle in the origin of Cenozoic continental basalts in east-central China . Lithos 2009 ; 110 : 305 – 26 . Google Scholar CrossRef Search ADS 92. Xu Z , Zhao ZF , Zheng YF . Slab-mantle interaction for thinning of cratonic lithospheric mantle in North China: geochemical evidence from Cenozoic continental basalts in central Shandong . Lithos 2012 ; 146 : 202 – 17 . Google Scholar CrossRef Search ADS 93. Xu YG , Zhang HH , Qiu HN et al. Oceanic crust components in continental basalts from Shuangliao, Northeast China: derived from the mantle transition zone? Chem Geol 2012 ; 328 : 168 – 84 . Google Scholar CrossRef Search ADS 94. Kuritani T , Ohtani E , Kimura JI . Intensive hydration of the mantle transition zone beneath China caused by ancient slab stagnation . Nat Geosci 2011 ; 4 : 713 – 6 . Google Scholar CrossRef Search ADS 95. Kuritani T , Kimura JI , Ohtani E et al. Transition zone origin of potassic basalts from Wudalianchi volcano, northeast China . Lithos 2013 ; 156 : 1 – 12 . Google Scholar CrossRef Search ADS 96. Michael PJ . Regionally distinctive sources of depleted MORB: evidence from trace-elements and H2O . Earth Planet Sc Lett 1995 ; 131 : 301 – 20 . Google Scholar CrossRef Search ADS 97. Dixon JE , Clague DA . Volatiles in basaltic glasses from Loihi seamount, Hawaii: evidence for a relatively dry plume component . J Petrol 2001 ; 42 : 627 – 54 . Google Scholar CrossRef Search ADS 98. Dixon JE , Leist L , Langmuir C et al. Recycled dehydrated lithosphere observed in plume-influenced mid-ocean-ridge basalt . Nature 2002 ; 420 : 385 – 9 . Google Scholar CrossRef Search ADS PubMed 99. Chen Y , Zhang Y , Graham D et al. Geochemistry of Cenozoic basalts and mantle xenoliths in Northeast China . Lithos 2007 ; 96 : 108 – 26 . Google Scholar CrossRef Search ADS 100. Hong LB , Zhang YH , Qian SP et al. Constraints from melt inclusions and their host olivines on the petrogenesis of Oligocene-Early Miocene Xindian basalts, Chifeng area, North China Craton . Contrib Mineral Petrol 2013 ; 165 : 305 – 26 . Google Scholar CrossRef Search ADS 101. Wang Y , Zhao ZF , Zheng YF et al. Geochemical constraints on the nature of mantle source for Cenozoic continental basalts in east-central China . Lithos 2011 ; 125 : 940 – 55 . Google Scholar CrossRef Search ADS 102. Chen H , Xia QK , Ingrin J . Water content of the Xiaogulihe ultrapotassic volcanic rocks, NE China: implications for the source of the potassium-rich component . Sci Bull . 2015 ; 60 : 1468 – 70 . Google Scholar CrossRef Search ADS 103. Chen H , Xia QK , Ingrin J et al. Heterogeneous source components of intraplate basalts from NE China induced by the ongoing Pacific slab subduction . Earth Planet Sci Lett 2017 ; 459 : 208 – 20 . Google Scholar CrossRef Search ADS 104. Liu J , Xia QK , Deloule E et al. Water content and oxygen isotopic composition of alkali basalts from the Taihang Mountains, China: recycled oceanic components in the mantle source . J Petrol 2015 ; 56 : 681 – 702 . Google Scholar CrossRef Search ADS 105. Liu SC , Xia QK , Choi SH et al. Continuous supply of recycled Pacific oceanic materials in the source of Cenozoic basalts in SE China: the Zhejiang case . Contrib Mineral Petrol 2016 ; 171 : 1 – 31 . Google Scholar CrossRef Search ADS 106. Wade JA , Plank T , Hauri EH et al. Prediction of magmatic water contents via measurement of H2O in clinopyroxene phenocrysts . Geology 2008 ; 36 : 799 – 802 . Google Scholar CrossRef Search ADS 107. Danyushevsky LV , Falloon TJ , Sobolev AV et al. The H2O content of basalt glasses from Southwest Pacific back-arc basins . Earth Planet Sci Lett 1993 ; 117 : 347 – 62 . Google Scholar CrossRef Search ADS 108. Danyushevsky LV , Eggins SM , Falloon TJ et al. H2O abundance in depleted tomoderately enrichedmid-ocean ridge magmas; part I: incompatible behaviour, implications for mantle storage, and origin of regional variations . J Petrol 2000 ; 41 : 1329 – 64 . Google Scholar CrossRef Search ADS 109. Dobson PF , Skogby H , Rossman GR . Water in boninite glass and coexisting orthopyroxene: concentration and partitioning . Contrib Mineral Petrol 1995 ; 118 : 414 – 9 . Google Scholar CrossRef Search ADS 110. Hochstaedter AG , Gill JB , Kusakabe M et al. Volcanism in the Sumisu Rift. I. Major element, volatile, and stable isotope geochemistry . Earth Planet Sci Lett 1990 ; 100 : 179 – 94 . Google Scholar CrossRef Search ADS 111. Sisson T , Layne G . H2O in basalt and basaltic andesite glass inclusions from four subduction-related volcanoes . Earth Planet Sci Lett 1993 ; 117 : 619 – 35 . Google Scholar CrossRef Search ADS 112. Stolper E , Newman S . The role of water in the petrogenesis of Mariana Trough magmas . Earth Planet Sci Lett 1994 ; 121 : 293 – 325 . Google Scholar CrossRef Search ADS 113. Wallace PJ . Water and partial melting in mantle plumes: inferences from the dissolved H2O concentrations of Hawaiian basaltic magmas . Geophys Res Lett 1998 ; 25 : 3639 – 42 . Google Scholar CrossRef Search ADS 114. Wallace PJ . Volatiles in subduction zonemagmas: concentrations and fluxes based on melt inclusion and volcanic gas data . J Volcanol Geotherm Res 2005 ; 140 : 217 – 40 . Google Scholar CrossRef Search ADS 115. Sun SS , McDonough WF . Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes . Geological Society, London, Special Publications 1989 ; 42 : 313 – 45 . Google Scholar CrossRef Search ADS 116. Cabral RA , Jackson MG , Koga KT et al. Volatile cycling of H2O, CO2, F, and Cl in the HIMU mantle: a new window provided by melt inclusions from oceanic hot spot lavas at Mangaia, Cook Islands . Geochem Geophys Geosyst 2014 ; 15 : 4445 – 67 . Google Scholar CrossRef Search ADS 117. Kendrick MA , Jackson MG , Kent AJ et al. Contrasting behaviours of CO2, S, H2O and halogens (F, Cl, Br, and I) in enriched-mantle melts from Pitcairn and Society seamounts . Chem Geol 2014 ; 370 : 69 – 81 . Google Scholar CrossRef Search ADS 118. Kendrick MA , Jackson MG , Hauri EH et al. The halogen (F, Cl, Br, I) and H2O systematics of Samoan lavas: assimilated-seawater, EM2 and high-3He/4He components . Earth Planet Sci Lett 2015 ; 410 : 197 – 209 . Google Scholar CrossRef Search ADS 119. Ruscitto DM , Wallace PJ , Cooper LB et al. Global variations in H2O/Ce. 2. Relationships to arc magma geochemistry and volatile fluxes . Geochem Geophys Geosyst 2012 ; 13 , 10.1029/2011GC003887 . 120. Wang XC , Wilde SA , Li QL et al. Continental flood basalts derived from the hydrous mantle transition zone . Nat Commun 2015 ; 6 : 7700 , 10.1038/ncomms8700 . Google Scholar CrossRef Search ADS PubMed 121. Chen H , Xia QK , Liu J et al. MTZ source of the Cenozoic basalts of eastern China constrained by H2O content . 2017 ; in preparation . 122. Workman RK , Hart SR . Major and trace element composition of the depleted MORB mantle (DMM) . Earth Planet Sc Lett 2005 ; 231 : 53 – 72 . Google Scholar CrossRef Search ADS 123. Zhi X , Song Y , Frey FA et al. Geochemistry of Hannuoba basalts, eastern China: constraints on the origin of continental alkalic and tholeiitic basalt . Chem Geol 1990 ; 88 : 1 – 33 . Google Scholar CrossRef Search ADS 124. Xu YG , Ma JL , Frey FA et al. Role of lithosphere–asthenosphere interaction in the genesis of Quaternary alkali and tholeiitic basalts from Datong, western North China Craton . Chem Geol 2005 ; 224 : 247 – 71 . Google Scholar CrossRef Search ADS 125. Sakuyama T , Tian W , Kimura JI et al. Melting of dehydrated oceanic crust from the stagnant slab and of the hydrated mantle transition zone: constraints from Cenozoic alkaline basalts in eastern China . Chem Geol 2013 ; 359 : 32 – 48 . Google Scholar CrossRef Search ADS 126. Xu YG . Recycled oceanic crust in the source of 90–40 Ma basalts in North and Northeast China: evidence, provenance and significance . Geochim Cosmochim Acta 2014 ; 143 : 49 – 67 . Google Scholar CrossRef Search ADS 127. Xu Z , Zheng YF , Zhao ZF et al. The hydrous properties of subcontinental lithospheric mantle: constraints from water content and hydrogen isotope composition of phenocrysts from Cenozoic continental basalt in North China . Geochim Cosmochim Acta 2014 ; 143 : 285 – 302 . Google Scholar CrossRef Search ADS 128. Bodnar RJ , Azbej T , Becker SP et al. Whole Earth geohydrologic cycle, from the clouds to the core: the distribution of water in the dynamic Earth system . In: Bickford ME (ed.). The web of geological sciences: advances, impacts, and interactions . Geol Soc Am Spec Paper 2013 ; 500 : 431 – 61 . Google Scholar CrossRef Search ADS 129. Panero WR , Pigott JS , Reaman DM et al. Dry (Mg, Fe)SiO3 perovskite in the Earth's lower mantle . J Geophys Res 2015 ; 120 : 894 – 908 . Google Scholar CrossRef Search ADS 130. Pearson DG , Brenker FE , Nestola F et al. Hydrous mantle transition zone indicated by ringwoodite included within diamond . Nature 2014 ; 507 : 221 – 4 . Google Scholar CrossRef Search ADS PubMed 131. Fukao Y , Obayashi M , Inoue H et al. Subducting slabs stagnant in the mantle transition zone . J Geophys Res 1992 ; 97 : 4809 – 22 . Google Scholar CrossRef Search ADS 132. Kelbert A , Schultz A , Egbert G . Global electromagnetic induction constraints on transition-zone water content variations . Nature 2009 ; 460 : 1003 – 6 . Google Scholar CrossRef Search ADS PubMed 133. Zhang J , Zhang HF , Ying JF et al. Contribution of subducted Pacific slab to Late Cretaceous mafic magmatism in Qingdao region, China: a petrological record . Isl Arc 2008 ; 17 : 231 – 41 . Google Scholar CrossRef Search ADS 134. Yang W , Teng FZ , Zhang HF et al. Magnesium isotopic systematics of continental basalts from the North China craton: implications for tracing subducted carbonate in the mantle . Chem Geol 2012 ; 328 : 185 – 94 . Google Scholar CrossRef Search ADS 135. Huang JL , Zhao DP . High-resolution mantle tomography of China and surrounding regions . J Geophys Res 2006 ; 111 ( B9 ), 10.1029/2005JB004066 . 136. Kuang YS , Wei X , Hong LB et al. Petrogenetic evaluation of the Laohutai basalts from North China Craton: melting of a two-component source during lithospheric thinning in the late Cretaceous-early Cenozoic . Lithos 2012 ; 154 : 68 – 82 . Google Scholar CrossRef Search ADS 137. Li HY , Huang XL , Guo H . Geochemistry of Cenozoic basalts from the Bohai Bay Basin: implications for a heterogeneous mantle source and lithospheric evolution beneath the eastern North China Craton . Lithos 2014 ; 196 : 54 – 66 . Google Scholar CrossRef Search ADS 138. Li HY , Xu YG , Ryan JG et al. Olivine and melt inclusion chemical constraints on the source of intracontinental basalts from the eastern North China Craton: discrimination of contributions from the subducted Pacific slab . Geochim Cosmochim Acta 2016 ; 178 : 1 – 19 . Google Scholar CrossRef Search ADS 139. Hawkesworth CJ , Hergt JM , Ellam RM et al. Element fluxes associated with subduction related magmatism . Philosophical Transactions of the Royal Society of London Series a-Mathematical Physical and Engineering Sciences 1991 ; 335 : 393 – 405 . Google Scholar CrossRef Search ADS 140. Tatsumi Y , Eggins S . Subduction Zone Magmatism . Oxford: Blackwell Science , 1995 , 211 . 141. Zheng YF . Fluid regime in continental subduction zones: petrological insights from ultrahigh-pressure metamorphic rocks . J Geol Soc 2009 ; 166 : 763 – 82 . Google Scholar CrossRef Search ADS 142. Zheng YF , Xia QX , Chen RX et al. Partial melting, fluid supercriticality and element mobility in ultrahigh-pressure metamorphic rocks during continental collision . Earth Sci Rev 2011 ; 107 : 342 – 74 . Google Scholar CrossRef Search ADS 143. Zhang JF , Jin ZM , Green HW et al. Hydroxyl in continental deep subduction zone: evidence from UHP eclogites of the Dabie Mountains . Chin Sci Bull 2001 ; 46 : 592 – 6 . Google Scholar CrossRef Search ADS 144. Zhang JF , Green HW , Bozhilov K et al. Faulting induced by precipitation of water at grain boundaries in hot subducting oceanic crust . Nature 2004 ; 428 : 633 – 6 . Google Scholar CrossRef Search ADS PubMed 145. Xia QK , Sheng YM , Yang XZ et al. Heterogeneity of water in garnets from UHP eclogites, eastern Dabieshan, China . Chem Geol 2005 ; 224 : 237 – 46 . Google Scholar CrossRef Search ADS 146. Sheng Y M , Xia Q K , Dallai L et al. H2O contents and D/H ratios of nominally anhydrous minerals from ultrahigh-pressure eclogites of the Dabie orogen, eastern China . Geochim Cosmochim Acta 2007 ; 71 : 2079 – 103 . Google Scholar CrossRef Search ADS 147. Chen RX , Zheng YF , Gong B et al. Origin of retrograde fluid in ultrahigh-pressure metamorphic rocks: constraints from mineral hydrogen isotope and water content changes in eclogite-gneiss transitions in the Sulu orogen . Geochim Cosmochim Acta 2007 ; 71 : 2299 – 325 . Google Scholar CrossRef Search ADS 148. Zhao ZF , Chen B , Zheng YF et al. Mineral oxygen isotope and hydroxyl content changes in ultrahigh-pressure eclogite-gneiss contacts from Chinese Continental Scientific Drilling Project cores . J Metamorph Geol 2007 ; 25 : 165 – 86 . Google Scholar CrossRef Search ADS 149. Zhao XD , Zhang ZM . Water in omphacites from UHP eclogites in the CCSD main hole . Acta Petrol Sin 2007 ; 22 : 2039 – 50 . 150. Aines , RD , Rossman GR . Water content of mantle garnets . Geology 1984 ; 12 : 720 – 3 . Google Scholar CrossRef Search ADS 151. Langer K , Rovarick E , Sobolev NV et al. Single-crystal spectra of garnets from diamondiferous high-pressure metamorphic rocks from Kazakhstan: indications for OH–, H2O, and FeTi charge transfer . Eur J Mineral 1993 ; 5 : 1091 – 100 . Google Scholar CrossRef Search ADS 152. Snyder GA , Taylor LA , Jerde EA et al. Archean mantle heterogeneity and the origin of diamondiferous eclogites, Siberia: evidence from stable isotopes and hydroxyl in garnet . Am Mineral 1995 ; 80 : 799 – 809 . Google Scholar CrossRef Search ADS 153. Matsyuk SS , Langer K , Hosch A . Hydroxyl defects in garnets from mantle xenoliths in kimberlites of the Siberian platform . Contrib Mineral Petrol 1998 ; 132 : 163 – 79 . Google Scholar CrossRef Search ADS 154. Ragozin AL , Karimovac AA , Litasov KD et al. Water content in minerals of mantle xenoliths from the Udachnaya pipe kimberlites (Yakutia) . Russ Geol Geophy 2014 ; 55 : 428 – 42 . Google Scholar CrossRef Search ADS 155. Bell DR , Rossman GR , Moore RO . Abundance and partitioning of OH in a high-pressure magmatic system: megacrysts from the Monastery kimberlite, South Africa . J Petrol 2004 ; 45 : 1539 – 64 . Google Scholar CrossRef Search ADS 156. Katayama I , Nakashima S , Yurimoto H . Water content in natural eclogite and implication for water transport into the deep upper mantle . Lithos 2006 ; 86 : 245 – 59 . Google Scholar CrossRef Search ADS 157. Skogby H , Janák M , Broska I . Water incorporation in omphacite: concentrations and compositional relations in ultrahigh-pressure eclogites from Pohorje, Eastern Alps . Eur J Mineral 2016 ; 28 : 631 – 9 . Google Scholar CrossRef Search ADS 158. Su W , You ZD , Cong BL et al. Cluster of water molecules in garnet from ultrahigh-pressure eclogite . Geology 2002 ; 30 : 611 – 4 . Google Scholar CrossRef Search ADS 159. Liu FL , Xu ZQ . Fluid inclusions hidden in coesite-bearing zircons in ultrahigh-pressure metamorphic rocks from southwestern Sulu terrane in eastern China . Chin Sci Bull 2004 ; 49 : 396 – 404 . Google Scholar CrossRef Search ADS 160. Zhang ZM , Shen K , Xiao YL et al. Mineral and fluid inclusions in zircon of UHP metamorphic rocks from the CCSD-main drill hole: a record of metamorphism and fluid activity . Lithos 2006 ; 92 : 378 – 98 . Google Scholar CrossRef Search ADS 161. Meng DW , Wu XL , Fan XY et al. Submicron-sized fluid inclusions and distribution of hydrous components in jadeite, quartz and symplectite-forming minerals from UHP jadeite-quartzite in the Dabie Mountains, China: TEM and FTIR investigation . App Geochem 2009 ; 24 : 517 – 26 . Google Scholar CrossRef Search ADS 162. Xiao YL , Zhang ZM , Hoefs J et al. Ultrahigh-pressure metamorphic rocks from the Chinese Continental Scientific Drilling Project. II. Oxygen isotope and fluid inclusion distributions through vertical sections . Contrib Mineral Petrol 2006 ; 152 : 443 – 58 . Google Scholar CrossRef Search ADS 163. Zheng Y , Zhao Z , Chen Y . Continental subduction channel processes: Plate interface interaction during continental collision . Chin Sci Bull 2013 ; 58 : 4371 – 7 . Google Scholar CrossRef Search ADS 164. Katayama I , Nakashima S . Hydroxyl in clinopyroxene from the deep subducted crust: evidence for H2O transport into the mantle . Am Mineral 2003 ; 88 : 229 – 34 . Google Scholar CrossRef Search ADS 165. Su W , Ji ZP , Ye K et al. Distribution of hydrous components in jadeite of the Dabie Mountains . Earth Planet Sci Lett 2004 ; 222 : 85 – 100 . Google Scholar CrossRef Search ADS 166. Hirschmann MM . Water, melting and the deep Earth water cycle . Annu Rev Earth Planet Sci 2006 ; 34 : 629 – 53 . Google Scholar CrossRef Search ADS 167. Li XP , Zheng YF , Wu YB et al. Low-T eclogite in the Dabie terrane of China: petrological and isotopic constraints on fluid activity and radiometric dating . Contrib Mineral Petrol 2004 ; 148 : 443 – 70 . Google Scholar CrossRef Search ADS 168. Guo S , Ye K , Chen Y et al. Fluid–rock interaction and element mobilization in UHP metabasalt: constraints from an omphacite–epidote vein and host eclogites in the Dabie orogen . Lithos 2012 ; 136–9 : 145 – 67 . Google Scholar CrossRef Search ADS 169. Lu R , Keppler H . Water solubility in pyrope to l00 kbar . Contrib Mineral Petrol 1997 ; 129 : 35 – 42 . Google Scholar CrossRef Search ADS 170. Withers AC , Wood BJ , Carroll MR . The OH contents of pyrope at high pressure . Chem Geol 1998 ; 147 : 161 – 71 . Google Scholar CrossRef Search ADS 171. Mosenfelder JL . Pressure dependence of hydroxyl solubility in coesite . Phys Chem Miner 2000 ; 27 : 610 – 7 . Google Scholar CrossRef Search ADS 172. Rauch M , Keppler H . Water solubility in orthopyroxene . Contrib Mineral Petrol 2002 ; 143 : 525 – 36 . Google Scholar CrossRef Search ADS 173. Koch-Müller M , Dera P , Fei YW et al. OH- in synthetic and natural coesite . Am Mineral 2003 ; 88 : 1436 – 45 . Google Scholar CrossRef Search ADS 174. Bromiley G . Solubility of hydrogen and ferric iron in rutile and TiO2 (II): implications for phase assemblages during ultrahigh-pressure metamorphism and for the stability of silica polymorphs in the lower mantle . Geophys Res Lett 2004 ; 31 : L04610 , 10.1029/2004GL019430(4) . Google Scholar CrossRef Search ADS 175. Mierdel K , Keppler H . The temperature dependence of water solubility in enstatite . Contrib Mineral Petrol 2004 ; 148 : 305 – 11 . Google Scholar CrossRef Search ADS 176. Yang XZ , McCammon C . Fe3+-rich augite and high electrical conductivity in the deep lithosphere . Geology 2012 ; 40 : 131 – 4 . Google Scholar CrossRef Search ADS 177. Bizimis M , Peslier AH . Water in Hawaiian garnet pyroxenites: implications for water heterogeneity in the mantle . Chem Geol 2015 ; 397 : 61 – 75 . Google Scholar CrossRef Search ADS © The Author(s) 2017. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

### Journal

National Science ReviewOxford University Press

Published: Feb 24, 2017

## You’re reading a free preview. Subscribe to read the entire article.

### DeepDyve is your personal research library

It’s your single place to instantly
that matters to you.

over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month ### Explore the DeepDyve Library ### Search Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly ### Organize Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place. ### Access Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals. ### Your journals are on DeepDyve Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more. All the latest content is available, no embargo periods. DeepDyve ### Freelancer DeepDyve ### Pro Price FREE$49/month
\$360/year

Save searches from
PubMed

Create lists to

Export lists, citations