Joint 52nd Northeastern Annual Section / 51st North-Central Annual Section Meeting - 2017

Paper No. 50-1
Presentation Time: 8:00 AM

A COMPARISON OF METHODS USED TO QUANTIFY CONDITIONS OF PARTIAL MELTING: A CASE STUDY FROM LINOSA, ITALY


WHITE, John C.1, MAY, Mitchell A2 and SIMPSON, Cassie E.1, (1)Department of Geosciences, Eastern Kentucky University, 521 Lancaster Ave, Roark 103, Richmond, KY 40475, (2)Freeport-McMoRan: Sierrita Operations, 6200 West Duval Mine Drive, Green Valley, AZ 85622, john.white@eku.edu

Linosa is a small (~6 km2) island situated in the Strait of Sicily Rift Zone (SSRZ), a northwest-southeast trending transtensional rift system situated on the Pelagian Block, the northern promontory of the African plate that represents the foreland domain of the Sicilian Apenninic-Maghrebian orogen. Linosa consists entirely of Na-alkaline mafic lavas and tuffs that erupted in three stages at 1070 ka, 700 ka, and 530 ka and created several coalescing cinder cone and maar volcanoes. Relatively primitive (MgO > 9 wt%) basaltic lavas occur on Linosa, which makes it an ideal setting to investigate the nature of the mantle and conditions of partial melting of the mantle in the SSRZ. In this study, we present and compare the results from four commonly used methods: (1) Estimation of the composition of primary magmas by iteratively “correcting” it for olivine fractionation until the recalculated basalt has an Mg# that has been experimental determined to be in equilibrium with mantle peridotite with an olivine composition of ~Fo90, followed by calculation of the average temperature and pressure of magma segregation; (2) Application of various trace- and major-element ratios that purport to identify the composition of the source of the magmas; (3) Rare-earth element inverse modelling using the INVMEL program to calculate melt curves to determine the degree of partial melting as well as the bottom and top of the melting column; and (4) Application of the pMELTS model to determine the conditions of isentropic partial melting for a variety of source compositions and potential temperatures over a similar range of pressures (3 GPa, the lower limit of pMELTS capability to 1.8 GPa, corresponding to the base of a ~60 km lithosphere. Preliminary results suggest that major- and trace-element ratios alone provide contradictory results. Results from the two inverse modelling techniques are consistent with each other, and suggest that magmatism at Linosa resulted from 2.0-2.5% partial melting of a lherzolite source consisting of ~65% depleted MORB mantle with a melting column extending from ~100 to 60 km (primarily in the spinel-garnet transition zone) with mantle potential temperatures of ~1425°C. Preliminary pMELTS results generally predict higher melt fractions (up to 5%) with elevated Na-concentrations, but work with this program continues.