The Minimum Potential Temperature of the Hawaiian Mantle is About 1420°C

1G.H. Gudfinnsson & 2D.C. Presnall

1Geophysical Laboratory, Carnegie Institution of Washington, 5251 Broad Branch Rd, NW, Washington, DC 20015-1305 USA, g.gudfinnsson@gl.ciw.edu

2Department of Geosciences, University of Texas at Dallas, P.O. Box 830688, Richardson, TX 75083-0688 USA, presnall@gl.ciw.edu
DCP alternate home page

Picritic glasses found in turbidite sands near the submarine part of Kilauea's East Rift Zone contain up to 15 wt.% MgO and are the most magnesian Hawaiian volcanic glasses reported to date (Clague et al., 1991, 1995). They have olivine phenocrysts as magnesian as Fo90.7, and when their compositions are plotted together on normative diagrams, they form a distinct olivine fractionation trend. Melt geothermometers indicate that the eruption temperatures of the picrite magmas were as high as about 1320¾C. On the assumption that these glasses represent primary melt compositions that coexisted with a lherzolite phase assemblage, the CMASNF geothermometer (Gudfinnsson & Presnall, 2001) yields a maximum temperature of generation of about 1480¾C, which corresponds roughly to a pressure of 2.5 GPa. This assumes that the melts were essentially free of H2O and CO2. However, both of these volatile components have the potential to lower significantly the solidus temperatures of mantle peridotite and alter the chemistry of primary melts. The approximately 0.4 wt.% H2O measured in the Hawaiian picrite glasses is probably below the saturation limit for H2O, and can be assumed to be close to the original H2O content of the picrite melts. The measured amount of CO2 in the glasses is low as most CO2 was probably lost by degassing at the time of eruption. The CO2 content of primary magmas at Kilauea has been determined as 0.7 wt.% (Gerlach & Graeber, 1985; Gerlach et al., 2001). Whereas the addition of CO2 tends to shift melts derived from peridotite toward greater alkalinity, the addition of H2O tends to move liquids toward the quartz normative side of the basalt tetrahedron. Results of CO2- and H2O-bearing melting experiments indicate that with the low amounts of H2O and CO2 expected in the primary melts of Kilauea their effect on the position of phase boundaries will be small. From phase relations involving melt in equilibrium with a garnet lherzolite phase assemblage in the system CaO-MgO-Al2O3-SiO2-CO2, we estimate that the solidus temperature of mantle lherzolite is lowered by about 10¾C for each 1 wt.% increase in the amount of CO2 in the melt. The effect of H2O is to lower the solidus temperature about 40¾C for each 1 wt.% increase in this component. This yields a minimum potential temperature (TP) for Hawaii of about 1420¾C, which is consistent with data indicating at most only a very slight increase of heat flow at Hawaii relative to Pacific crust of the same age (Stein & Stein, 1993). This TP is 140-160¾C higher than petrological estimates of the average TP of the MORB source (McKenzie & Bickle, 1988; Presnall et al., 2002). Our data do not constrain the upper limit of TP at Hawaii.