Depth of curie temperature in continental shields: a compositional boundary?

Abstract

MUCH attention is being given to estimating the depth to the bottom of the crustal magnetised layer. The latter can be interpreted as the surface where ferrimagnetic minerals, which are responsible for the magnetisation of crustal rocks, reach their Curie temperatures 1·5. The depth of this surface is therefore temperature dependent and it can be used to locate thermal anomalies within the continental crust. One application would be in the search for shallow intercrustal heat sources (such as large magmatic reservoirs or cooling plutons) which would be of economic interest as natural energy sources^3,6. In all these works the bottom of the crustal magnitised layer is assumed to correspond to an isotherm whose temperature is approximately the Curie temperature of pure magnetite (580 °C). Haggerty⁷ cautioned against assuming a single Curie temperature for the continental crust and pointed out that Curie temperatures as low as 300 °C may exist in the crust because of the low temperature oxidation of titaniferous magnetite. Furthermore, he suggested that depths of Curie temperatures greater than 20 km may be due to serpentinisation of mafic and ultramafic components of the continental lower crust with consequent formation of Fe–Co–Ni alloys with Curie temperatures in the range 620–1,100 °C. We show here that depths of Curie temperatures computed for continental areas characterised by normal and low heat flow cannot be interpreted as isothermal surfaces with temperatures at or above 580 °C. We suggest that the bottom of the crustal magnetised layer may correspond to vertical compositional changes within the crust.