Lithospheric thickness controlled compositional variations in potassic basalts of Northeast China by melt‐rock interactions

Abstract

Melt‐rock interaction is a common mantle process; however, it remains unclear how this process affects the composition of potassic basalt. Here we present a case study to highlight the link between compositional variations in the potassic basalts and melt‐rock interaction in cold lithosphere. Cenozoic potassic basalts in Northeast China are strongly enriched in incompatible elements and show EM1‐type Sr–Nd–Pb isotopes, suggesting an enriched mantle source. These rocks show good correlations between 87Sr/86Sr and K2O/Na2O and Rb/Nb. Notably, these ratios decrease with increasing lithospheric thickness, which may reflect melt‐lithosphere interaction. Phlogopite precipitated when potassic melts passed through the lithospheric mantle, and K and Rb contents of the residual melts decreased over time. The thicker the lithosphere, the greater the loss of K and Rb from the magma. Therefore, the compositions of potassic basalts were controlled by both their enriched sources and reactions with lithospheric mantle.

The role of melt‐rock interaction in the formation of Quaternary high‐MgO potassic basalt from the Greater Khingan Range, northeast China

“…However, mantle xenoliths from the Keluo potassic basalts have moderately depleted Sr‐Nd‐Hf isotopic composition [ Zhang et al ., ], and they therefore cannot be the source of these potassic and ultrapotassic lavas (Figures a and b). Furthermore, Liu et al [] have highlighted that the Sr‐Nd isotopic composition becomes less enriched, MgO content increases, and K 2 O/Na 2 O and Rb/Nb decrease from the Erkeshan to the Nuominhe potassic basalts with increasing lithospheric thickness below the individual volcanic fields, which also indicates that the lithospheric mantle is not the direct source of the primary potassium‐enriched melts and that it modified their composition by melt‐peridotite interactions. Therefore, the ultimate (primary) mantle source of the Cenozoic potassic (and ultrapotassic) basalts in northeast China, including that of the Nuominhe high‐MgO potassic basalt, is located within the asthenosphere.…”

“…Experimental studies have confirmed that reaction between a siliceous melt and peridotite produces silica‐undersaturated alkali melts owing to the formation of orthopyroxene and/or garnet at the expense of olivine [ Mallik and Dasgupta , , , ], where the degree of alkalinity and silica undersaturation of the resulting melt increase with increasing system CO 2 content [ Mallik and Dasgupta , ]. Despite abundant evidence that melt‐rock interaction occurs in the shallow mantle and contributes to the compositional diversity of intraplate basalt [ Xu et al ., ; Søager and Holm , ; Zeng et al ., ; Liu et al ., ], it is still not well understood how these reactions and thus the erupted basalt compositions spatially and temporally evolve, which we do in this contribution.…”

“…Our previous studies have shown that 87 Sr/ 86 Sr, K 2 O/Na 2 O, and Rb/Nb of the potassic rocks are well correlated and that they decrease with increasing lithospheric thickness below individual volcanic centers [ Liu et al ., ]. This correlation is well explained by variable degrees of interaction between the primary potassic melts and a cold lithospheric mantle, where reaction efficacy enhanced with increasing lithospheric thickness [ Liu et al ., ]. The Nuominhe volcanic rocks, which are underlain by the thickest lithosphere of all investigated potassic volcanic fields (Figure b), show the strongest geochemical evidence for melt‐rock interaction.…”

Hot, volatile‐poor, and oxidized magmatism above the stagnant Pacific plate in Eastern China in the Cenozoic

“…Trace to minor amounts of H 2 O and CO 2 , as we infer them for the natural systems, were very likely present in the experiments (they are inevitable contaminants; e.g., Spandler et al, ), while higher volatile contents (i.e., >1 wt% combined) can be ruled out for the experimental systems and thus also for the sources of our investigated natural systems. Spatiotemporally associated sodic basalts are interpreted to derive from depleted mantle sources with minor contributions from enriched components, while spatiotemporally associated potassic basalts erupted at Wudalianchi, Nuominhe, and Xiaogulihe, have long been interpreted to comprise partial melts derived from metasedimentary eclogite with contributions of depleted mantle through reaction or metasomatism (Liu et al, ; J.‐Q. Liu et al, ; Wang et al, ).…”

“…Potassic basaltic, basaltic andesitic and andesitic magmatism, in contrast, is abundant above or close to the present‐day edge of the stagnant Pacific plate and above some of the major low‐ V P , low V S anomalies (Figure a; Fan & Hooper, ; Ho et al, ; Kuritani et al, , 2013; Lim et al, ; Liu et al, ; Wang et al, ; Xu et al, ). Most studies conclude that the potassic basalts derive from enriched mantle (EM1‐type) which contains subducted sedimentary eclogite, where melts formed during upwelling in the asthenosphere and subsequently reacted with the lithosphere (Kuritani et al, ; Liu et al, , J.‐Q. Liu et al, ; Wang et al, ; Xu et al, ).…”