High water content in primitive continental flood basalts

Qun-Ke Xia¹, Yao Bi², Pei Li¹, Wei Tian³, Xun Wei⁴, Han-Lin Chen¹

¹School of Earth Sciences, Zhejiang University, Hangzhou, China; qkxia@zju.edu.cn ; peili@zju.edu.cn ; hlchen@zju.edu.cn
²School of Earth and Space Sciences, University of Science and Technology of China, Hefei, China; biyao@mail.ustc.edu.cn
³School of Earth and Space Sciences, Peking University, Beijing, China; davidtian@pku.edu.cn
⁴Key Laboratory of Marine Geology and Environment, Institute of Oceanology, Chinese Academy of Sciences, Qingdao, China; weixun@yeah.net

This webpage is a summary of Xia, Qun-Ke, Yao Bi, Pei Li, Wei Tian, Xun Wei & Han-Lin Chen, High water content in primitive continental flood basalts, Scientific Reports 6, Article number: 25416 (2016) doi:10.1038/srep25416

Introduction

In principle, the generation of continental flood basalts (CFB) requires abnormally high temperature, extended decompression, mafic rock in the mantle source, or H₂O and/or CO₂ in the mantle source (Campbell, 2001; Sobolev et al., 2011). These four factors are not mutually exclusive, and it is likely that several or all factors contribute together to generate CFB. Elevated H₂O and/or CO₂ content allows melting to start at greater depth and enlarges the whole melting region, consequently boosting melt volume. The CO₂ content is much less than the H₂O content in the mantle (Marty, 2012), and CO₂ lowers the solidus less than H₂O (Green et al., 2015). H₂O is thus expected to be more important to the genesis of CFB. High temperature, decreased pressure and mafic source lithology have been discussed for three decades (White & McKenzie, 1989; Campbell & Griffiths, 1990; Anderson, 1995, 2005; Foulger, 2002; Sobolev et al., 2011; Hole et al., 2015), but less is known about H₂O.

Measuring H₂O content in basaltic magma using clinopyroxene phenocrysts

Because the partition coefficient of H₂O between clinopyroxene (cpx) and basaltic melt is well constrained experimentally (Hauri et al., 2006; O’Leary et al., 2010), early crystallized cpx phenocrysts can be used to infer the initial H₂O content of basaltic magma (Wade et al., 2008). Recent studies have shown that if the OH bands in cpx are approximately 3640–3620, 3540–3520 and 3470–3450 cm^-1 and the maximum linear absorption is <0.3, Fourier Transform Infrared Spectroscopy (FTIR) analysis with an unpolarized beam conducted on a single cpx grain can allow estimation of H₂O content with a 20% deviation at most from the result obtained by polarized IR analysis along the principal optical directions of cpx. The latter method is theoretically most accurate but difficult to apply to cpx phenocrysts in basalts because of their small sizes and the presence of cracks (Xia et al., 2013; Liu et al., 2015). Up to now, this has been the easiest way of measuring H₂O content in cpx phenocrysts. For most augite and diopside phenocrysts in basalts, the H₂O content can be obtained with an uncertainty of <30% and that of basaltic magma with an uncertainty of <40% (Liu et al., 2015).

Results and discussion

The Tarim basin in NW China is an early Permian large igneous province (LIP) with > 200,000 km² of flood basalts (Figure 1). Many cpx macrocrysts (1-15 mm in size) were found in a basaltic dyke that crosscuts to the Early Permian (~280 Ma) Xiaohaizi wehrlite intrusion in the western Tarim basin. They commonly have a high-Mg core (Mg# 80-89) and a thin low-Mg rim (Mg# down to 70) resulting from interaction with the host basalt (Wei et al., 2015). The preservation of the high-Mg character and depleted Sr-Nd isotope compositions (⁸⁷Sr/⁸⁶Sr_i = 0.7035-0.7037, and εNd_i = 4.5-4.8) suggest that the cores of these cpx macrocrysts may have recorded the composition of the primitive Tarim basaltic melts, with little crustal contamination (Wei et al., 2015).

Figure 1: (a) Location of the Tarim LIP and Xiaohaizi intrusion. (b) Clinopyroxene macrocrysts in the Xiaohaizi dyke. (c) Backscattered electron image of a typical clinopyroxene macrocryst  with a gray core and a thin, bright rim. Red squares are five spots used for FTIR and EPMA analyses, and the inserted figure shows the IR spectra of five spots that are the same both in band position and height.

The chemical composition and H₂O content in 10 cpx macrocrysts were obtained by an electron probe micro-analyzer (EPMA) and an FTIR. Only the core of each grain was measured in order to retrieve information about the initial and primitive basaltic melts. Five to seven clean analyzed spots in the core area were selected to run EPMA and FTIR for each grain. In individual grains they show the same chemical compositions and IR spectra (Figure 1). The average values of the analyzed spots of each grain were thus used to represent the element and H₂O contents of that grain. The cpx Mg# values are 85.2 to 87.8, corresponding to a Mg# of ~70 for the equilibrated basaltic melts using the experimental Mg-Fe partition coefficient (0.34 ± 0.04; Kinzler, 1997). This suggests that the analyzed cpx grains crystallized from a nearly primary basaltic source (Frey et al., 1978), in agreement with the trace-element and Sr-Nd isotope characteristics.

The H₂O contents of 10 cpx grains are 300-550 wt. ppm (Average: 384 ± 83 wt. ppm), and the calculated H₂O contents of the equilibrated basaltic melts are 3.69 wt.% to 6.61 wt.% (Average: 4.82 ± 1.00 wt.%) (Figure 2). Although the uncertainty is up to 40%, such an H₂O content is apparently higher than those of MORB, OIB and back-arc basin basalts (0.2 - 2.0 wt.%). Instead, it falls in the range of arc magmas (2.0 - 8.0 wt.%; Figure 2).

Figure 2: Comparison of the initial H₂O contents of the Tarim primitive basalts with those of MORB, OIB, back-arc basin basalts (BABB) and island-arc basalts (IAB).

The arc-like H₂O content of the early Permian Tarim primary basaltic melts suggests addition of water from subduction-related processes. However, the Tarim basalts do not display arc-like geochemical signatures (i.e., LILE-enrichment and HFSE-depletion; Zhou et al., 2009; Zhang et al., 2010; Yu et al., 2011; Li et al., 2012). This suggests that the extra water in the source of the Tarim basalts was not from fluids released from subducting plates but instead came from dehydrated plates stagnated in the deep Earth.

If we consider:

then ~2500-5000 ppm wt. H₂O in the source can produce 5% H₂O in basaltic melts, regardless of the melting model used (batch or fractional).

The upper mantle can accommodate several hundred ppm (wt.) of H₂O (Hirschmann et al., 2005; Ohtani, 2015) and the lower mantle contains much less (Panero et al., 2015). Only the mantle transition zone (MTZ) can contain >1 wt.% H₂O (Kohlstedt et al., 1996; Pearson et al., 2014). Several thousand ppm (wt.) of H₂O in the source of the Tarim basalts thus probably arises from the MTZ where subducted plates stagnate and provide water (Fukao et al., 1992). If so, the classic core-mantle boundary-derived plume model (Campbell & Griffiths, 1990; Xu et al., 2014) cannot be applied to the Tarim LIP.

In conclusion, the high water content in the primary Early Permian Tarim basalts provides clear evidence that water, in addition to temperature, pressure and source lithology, plays an important role in the generation of continental flood basalts. Furthermore, when high water content is considered, abnormally high temperature and extended decompression, two critical factors in the widely accepted mantle plume model (Campbell, 2001), are not always required in the generation of CFB and LIPs.

Acknowledgements

Financial support was provided by National Natural Science Foundation of China (Nos. 41225005 and 41330207).

References