Discussion Implications for the fluid The shallow axial crust

The mid-crustal low resistivity anomaly

The most significant result of this study is the identification of a substantial low resistivity zone beneath the axis of the AVR segment at 57°45 ' N on the Reykjanes Ridge. The anomaly constrained by the CSEM data is consistent in position and geometry with the low velocity zone detected by the wide angle seismic component of the integrated geophysical study (Navin etal, this issue) beneath the same AVR segment. It is also consistent with the high crustal conductance required to explain the MT data (Heinson , this issue). Resistivities in the crust which are as low as those observed, in a zone which is coincident with a significant reduction in seismic velocity, can be most easily explained if there is melt in the crust.

Laboratory measurements (Waff & Weill, 1975; Tyburczy & Waff, 1983) show that the resistivity of pure basaltic melt is in the range 1 to 0.1 , for temperatures between approximately 1200°C and 1500°C. In partially molten regions the resistivity depends much more strongly on the resistivity of the molten phase than that of the solid, and resistivity of melt is affected much more by variations in temperature than by other parameters such as pressure or oxygen fugacity. Since the resistivity of molten basalt increases with temperature, a low bulk resistivity in a medium can be achieved either by a large melt fraction and a relatively low temperature, or by a lower melt fraction and higher temperature (Shankland & Waff, 1977; Tyburczy & Waff, 1983). As before, this makes inference of the melt fraction from a resistivity measurement ambiguous.

However, for a given observation of electrical resistivity, the possible range of temperatures and melt fractions may be examined. It has been shown experimentally that for melt fractions as low as 3% , the melt forms an interconnected network along the grain boundaries (Tyburczy & Waff, 1983). The upper HS bound (equation 5) can therefore be applied to obtain a lower bound on the possible melt fraction. Figure 10 shows curves of constant effective resistivity in a two phase medium consisting of a basaltic melt and a solid phase, calculated by Shankland & Waff (1977), using the experimentally determined variation of basaltic melt resistivity with temperature of Waff & Weill (1975). The resistivity of the solid phase was taken to be much higher than that of the melt so that it contributed relatively little to the conduction process. The results were calculated at zero pressure. Tyburczy & Waff (1983) showed that the resistivity of basaltic melt increases slightly with pressure for pressures up to 5-8 kbar. The pressure at 2 km depth in the crust beneath 2 km of seawater is approximately 1 kbar and therefore including the pressure effect would increase the melt fraction required to explain an observed resistivity. The low pressure curves are appropriate for minimum melt fraction calculations.

Geologically reasonable estimates of the melt fraction required to explain an observed resistivity may be obtained by assuming that the basaltic melt in a crustal magma chamber is at a temperature between its solidus and liquidus. The solidus and liquidus temperatures of basalt, determined experimentally by Presnall etal (1972) at atmospheric pressure are plotted in Figure 10 . Bender etal (1978) studied the melting of basalts collected from the median valley of the Mid-Atlantic Ridge at pressures from atmospheric to 15 kbar. The liquidus temperature rises by approximately 20 °C between 0 and 8 kbar. The error introduced to the inferred melt fraction by assuming the solidus and liquidus at atmospheric pressure is therefore small.

The upper bound on the resistivity of the sub-axial anomaly is 2.5 , constrained by the CSEM data. A lower bound of 1.5 is given by the MT results (Heinson etal, this issue), which constrain the maximum crustal conductance. Figure 10 shows that the melt fraction is constrained to be at least 20%, and may be as high as 45%. These values are much larger than estimates of a few percent in the low velocity zones detected seismically at the East Pacific Rise (Caress etal, 1992; Wilcock etal, 1995), suggesting a significant difference in melt content between the magma chambers detected at fast and slow spreading rates.

Discussion Implications for the fluid The shallow axial crust

Go to first page