The effect of in-situ metasomatism on the electrical resistivity of the lower crust

Abstract

Improved estimates of composition and temperature in the lower crust are vital in characterising crustal rheology and better understanding tectonic processes. Fluids introduced into the lower crust during collisional events leave behind large domains of metasomatised crust which are long-lived and should be observable via magnetotelluric sounding. With this in mind, the electrical conductivity of a subduction-related eclogite and its anhydrous, anorthositic granulite protolith were experimentally determined at atmospheric pressure and lower crustal temperatures (773–1123 K). Complex impedance spectra were collected between 20 Hz and 2 MHz while oxygen fugacity was controlled about the quartz-fayalite-magnetite (QFM) buffer using a CO:CO2 gas mix. At <1000 K, both samples are characterised by impurity conduction with low observed activation enthalpies of 0.48 ± 0.03 eV and 0.39 ± 0.02 eV for the granulite and eclogite respectively. High temperature data from the granulite (>1000 K) indicates a switch to small polarons with an activation enthalpy of 1.02 ± 0.02 eV. Results indicate that conversion of Fe-poor mafic granulite to eclogite does not enhance conductivity at lower crustal temperatures. Comparisons with previous experiments reveal a much stronger relationship between conductivity and FeOT than previously anticipated. Modelling indicates that an increase in FeOT of 2 wt% is equivalent to a temperature increase of 100 K and reasonable variations in FeOT in the lower crust may account for ~4 orders of magnitude variation in conductivity. Elevated FeOT up to 20 wt% may enhance conductivity to ~100 S m− 1 at a moderate 873 K. A system of linear equations is derived to estimate iron content, electrical conductivity or temperature within dry, mafic lower crust. For the temperature range 773–973 K, conductivity may be predicted to within 0.16 log units using log σ(T,FeOT) = 0.006 T + 0.3FeOT − 11.35. Under the same conditions, FeOT may be predicted to within 0.51 wt% using FeOT (T,σ) = 37–0.0193 T + 3.24 log σ. Finally, lower crustal temperatures may be estimated to within 24 K using T(FeOT,σ) = 1728 − 42FeOT + 139 log σ.