Journal of Conference Abstracts

Water and Carbon Dioxide in Mid-Atlantic Ridge Glasses (22°­41°N): Implications for the Role of Water in the Generation of MORB

Jacqueline Eaby Dixon Rosenstiel School of Marine & Atm. Sci., Univ. of Miami, 4600 Rickenbacker Causeway,

Miami, FL 33149, USA

jdixon@rsmas.miami.edu

Charles H. Langmuir Lamont-Doherty Earth Observatory, University of Columbia, Palisades,

NY 10964-8000, USA

Shannon Horan Rosenstiel School of Marine & Atm. Sci., Univ. of Miami, 4600 Rickenbacker Causeway,

Miami, FL 33149, USA

Concentrations of dissolved H₂O (0.10-1.29 wt%), CO₂ (25-325 ppm), and Cl (50-1022 ppm) were determined in glasses from 22°-41°N on the MAR using infrared spectroscopy and electron microprobe. Most samples are supersaturated with respect to a mixed CO₂-H₂O vapor with calculated vapor saturation pressures ranging from 1-2x the eruption pressures (Figs. 1 and 2). Samples erupted on the summit of the Azores Platform have high H₂O contents (0.8-1.29 wt%) and reached saturation for their depth of eruption. High water contents may increase the nucleation and growth rates of bubbles, thus faciliating attainment of vapor-melt equilibrium on eruptive timescales.

Fig. 1: Pressures of vapor saturation calculated according to procedure in Dixon and Stolper (1995).

Fig. 2: Filled diamonds - Kane to the Hayes Fracture Zones (22°-33.6°N). Filled circles - Hayes to Oceanographer Fracture Zones (33.6°-35°N). Open triangles - Oceanographer Anomaly (~35.2°N). Open squares - Azores Platform (35°-41°N). Open squares with pluses - MAR (35°-62°N; Michael, 1995). Crosses - Pacific and South Atlantic MORB (Michael, 1995). Samples AII127-D9 (dredged at 1657 m on the summit of an off-axis seamount) and AII127-D17 (dredged at 926 m on a summit cone on the Azores Platform) have relatively water rich equilibrium vapor compositions and have probably degassed significant amounts of H₂O.

Water concentrations are similar to previously published values for this region (Kingsley, 1989; Michael, 1995). In general our results confirm conclusions of previous studies including: 1) H₂O behaves incompatibly during melting and crystallization with a relative order of incompatibility of K₂O>H₂O~Ce>P₂O₅, therefore H₂O/Ce can be used as an indicator of mantle enrichment or depletion in H₂O (Michael, 1988, 1995; Dixon et al., 1988); 2) Values of H₂O/Ce range from 190-400 consistent with the conclusion that the entire North Atlantic is enriched in H₂O with respect to the Pacific and South Atlantic (Michael, 1995); 3) Primary basalts erupted on the Azores platform contain about twice as much H₂O as depleted MORB (Schilling et al., 1983; Kingsley, 1989; Michael, 1995).

Chlorine concentrations are similar to previously published values (Schilling et al., 1980). Plots of Cl/K₂O, and Cl/P₂O₅ versus (La/Sm)_n or Ba/TiO₂ show smooth trends that indicate a relative order of incompatibility of K₂O>Cl>P₂O₅. Our data indicate that Cl behaves more compatibly than determined by Schilling et al. (1980).

Whereas previous work focused on regional variations in mantle H₂O concentrations, the close sample spacing in this study allows us to investigate exciting new local scale variations (Figs. 3 - 6). In particular, the section of the MAR between the Kane and Hayes Fracture Zones (24°-34°N) is a normal, depleted mid-ocean ridge characterized by low (La/Sm)_n (0.43-0.68) and low concentrations of K₂O (0.03-0.16) P₂O₅ (0.08-0.18), but with relatively high H₂O (0.12-0.35). The extent of melting (as inferred from Na₈) and the crustal thickness (as inferred from depth) increase from south to north accompanied by decreasing (La/Sm)_n and concentrations of most incompatible elements (e.g., P₂O₅, K₂O and Ce). In contrast, concentrations of H₂O and Ba, though scattered, remain essentially constant. H₂O/Ce ratios increase from south to north with a range (200-340) almost as large as that found on the Azores Platform (200-400). H₂O and K₂O concentrations are decoupled, consistent with observations for primary MORB in melt inclusions (Sobolev and Chaussidon, 1996). We suggest that heterogeneity in mantle water concentration may affect the extent of melting along this "normal" depleted segment of the MAR.

North of the Hayes Fracture Zone, concentrations of highly incompatible elements like K₂O and Ba increase steadily northward. In contrast, concentrations of H₂O are markedly lower between Hayes to Oceanographer and then increase substantially over the Azores platform, where the most enriched and fractionated basalts have water contents as high as 1.3%. Water correlates well with other incompatible elements such as P, Cl and U over this region. The ratio of water to incompatible elements is lower than observed in back-arc basin basalts (BABB). Therefore if a hydrous phase is contributing to the heterogeneity sampled by the Azores Platform, it has a different composition than that influencing BABB.

One of the major aspects of the effect of water on melting during adiabatic ascent is the increase in melt below the dry solidus as the amount of water increases. This can cause significant changes in melt composition as well as melt volume. The effect of water on the mean extent of melting as determined by isobaric experiments does not adequately describe the total effect of water on the melting regime. We are in the process of modelling the effect of water on melting during adiabatic ascent in order to separate the relative roles of mantle temperature and volatile contents on extent of melting and basalt chemistry.

Figs 3 - 6: K - Kane FZ, A - Atlantis FZ, H - Hayes FZ, O - Oceanographer FZ. Symbols-same as in Fig. 2.

References

Dixon, J.E. & Stolper, E.M., J. Pet. 36, 1633-1646 (1995).

Dixon, J.E., Stolper, E.M. & Delaney, J.R., Earth Planet. Sci. Lett. 90, 87-104 (1988).

Kingsley, R., Master's Thesis, University of Rhode Island, 146 pp. (1989).

Michael, P.M., Geochim. Cosmochim. Acta 52, 555-566 (1988).

Michael, P.M., Earth Planet. Sci. Lett. 131, 301-320 (1995).

Schilling, J.-G., Bergeron, M.B., & Evans, R., Phil. Trans. R. Soc. Lond. A 297, 147-178 (1980).

Schilling, J.-G., Zajac, M., Evans, R., Johnston, T., White, W., Devine, J.D., & Kingsley, R., Am. J. Sci. 283, 510-586 (1983).

Sobolev, A.V. & Chaussidon, M., Earth Planet. Sci. Lett. 137, 45-55 (1996).

Stolper, E.M. & Holloway, J.R., Earth Planet. Sci. Lett. 87, 397-408 (1988).

Stolper, E.M. & Newman, S., Earth Planet. Sci. Lett. 121, 293-325 (1994).