Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Durham University

Department of Earth Sciences

Profile

Publication details for Professor Yaoling Niu

Niu, Yaoling & Batiza, R. (1994). Magmatic processes at a slow spreading ridge segment: 26°S Mid-Atlantic Ridge. Journal of Geophysical Research: Solid Earth 99(B10): 19719-19740.

Author(s) from Durham

Abstract

We present new data on the mineralogy, and major and trace element compositions of lavas dredged along a segment of the Mid-Atlantic Ridge at ∼26°S. This segment is bounded on the north by the Rio Grande transform and on the south by the Moore discontinuity and has an along-axis, central high rising to ∼2600 m near the middle of the segment. The segment is well-studied and the dredges are spaced ∼7 km apart. The lavas are exclusively normal mid-ocean ridge basalt (N-MORB), with a limited range of MgO (8.62–6.55 wt %). Petrographically, the lavas are dominated by low-pressure mineral assemblages, with two distinct crystallization sequences: Type I basalts have pl [RIGHTWARDS ARROW] pl + ol [RIGHTWARDS ARROW] pl + ol + cpx and come mainly from the interior of the segment; type II basalts, mainly from near the offsets, have ol [RIGHTWARDS ARROW] ol + pl [RIGHTWARDS ARROW] ol + pl + cpx. Some of the lavas contain rare anorthitic megacrysts and one sample contains a small gabbroic inclusion. The mineralogy and mineral chemistry are typical of N-MORB lavas elsewhere. Type I lavas have higher CaO and lower FeO, TiO2, and Na2O than type II, but similar A2O3 contents. Type I lavas define curvilinear trends on MgO variation diagrams and represent diverse parental melts that have undergone variable extents of low-pressure fractionation. In contrast, type II lavas exhibit less fractionation and more diversity of inferred parental melt compositions. Fractionation of all the lavas appears to have been mainly at low pressure, though we cannot rule out the possibility of some higher-pressure, polybaric crystallization. Except for lavas near the offsets, there is a rough correlation of lava chemistry and axial depth, defining along-axis “W” and “M” patterns. We infer that the center of the segment has higher magma supply than the ends and that magma chambers in the center are larger and more like those at the East Pacific Rise than those near offsets. Overall, the segment is fed by diverse magma types with different melting histories that retain their chemical distinctiveness all the way to the surface. Ni, [La/Sm]N, and other trace elements also display “W” patterns, as do isotope ratios, but the patterns are noisier than those of the major elements. We infer that the mantle source of the lavas is heterogeneous but its range of variability in major and trace elements and isotope ratios appears to be small. The rare earth elements and major elements normalized to 8 wt % MgO exhibit U-shaped along-axis variation patterns. Calculations show that the along-axis initial depth of melting is constant at ∼50 km but that the final depth of melting varies from ∼35 km near offsets to ∼28 km in the center. The mean extent of melting varies from 15% near offsets to 18% in the center and correlates with values of the mantle Bouguer anomaly, suggesting that the gravity pattern is of shallow feature (<50 km). We suggest that the melting patterns and density structure result from the combined effect of along-axis variation in mantle upwelling and melt production and the cold edges near offsets. The 26°S segment exhibits the local trend of chemical variation. New chemical evidence (Ca, Ti, and Ni data) strengthens the hypothesis that the local trend at slow spreading ridges is due to the melting reaction: melt A + pyroxene [RIGHTWARDS DOUBLE ARROW] melt B + olivine. We suggest that this reaction, occurring in ascending and melting diapirs, is an important process at slow spreading ridges, consistent with gravity and modeling studies.