The role of water in generating the calc-alkaline trend: New volatile data for aleutian magmas and a new tholeiitic index

The origin of tholeiitic (TH) versus calc-alkaline (CA) magmatic trends has long been debated. Part of the problem stems from the lack of a quantitative measure for the way in which a magma evolves. Recognizing that the salient feature in many TH–CA discrimination diagrams is enrichment in Fe during magma evolution, we have developed a quantitative index of Fe enrichment, the Tholeiitic Index (THI): THI = Fe_4·0/Fe_8·0, where Fe_4·0 is the average FeO* concentration of samples with 4 ± 1 wt % MgO, and Fe_8·0 is the average FeO* at 8 ± 1 wt % MgO. Magmas with THI > 1 have enriched in FeO* during differentiation from basalts to andesites and are tholeiitic; magmas with THI < 1 are calc-alkaline. Most subduction zone volcanism is CA, but to varying extents; the THI expresses the continuum of Fe enrichment observed in magmatic suites in all tectonic settings. To test various controls on the development of CA trends, we present new magmatic water measurements in melt inclusions from eight volcanoes from the Aleutian volcanic arc (Augustine, Emmons, Shishaldin, Akutan, Unalaska, Okmok, Seguam, and Korovin). Least degassed H₂O contents vary from ∼2 wt % (Shishaldin) to >7 wt % (Augustine), spanning the global range in arc mafic magmas. Within the Aleutian data, H₂O correlates negatively with THI, from strongly calc-alkaline (Augustine, THI = 0·65) to moderately tholeiitic (Shishaldin, THI = 1·16). The relationship between THI and magmatic water is maintained when data are included from additional arc volcanoes, back-arc basins, ocean islands, and mid-ocean ridge basalts (MORBs), supporting a dominant role of magmatic water in generating CA trends. An effective break between TH and CA trends occurs at ∼2 wt % H₂O. Both pMELTs calculations and laboratory experiments demonstrate that the observed co-variation of H₂O and THI in arcs can be generated by the effect of H₂O on the suppression of plagioclase and the relative enhancement of Fe-oxides on the liquid line of descent. The full THI–H₂O array requires an increase in fO₂ with H₂O, from ≤FMQ (where FMQ is the fayalite–magnetite–quartz buffer) in MORB to ∼ΔFMQ +0·5 to +2 in arcs, consistent with inferences from measured Fe and S species in glasses and melt inclusions. A curve fit to the data, H₂O (wt % ± 1·2) = exp[(1·26 – THI)/0·32], may provide a useful tool for estimating the H₂O content of magmas that are inaccessible to melt inclusion study.