Petrogenesis of monzonoritic dykes in the Egersund-Ogna anorthosite (Rogaland, S.W. Norway): trace elements and isotopic (Sr, Pb) constraints

Abstract

Two major monzonoritic dykes occur in the Egersund-Ogna anorthositic massif (S.W. Norway): the Lomland dyke, which varies from norite to monzonite, and the Vettaland ferronoritic dyke. They are characterized by high Fe, Ti and P contents, low SiO2, variable K2O (0.5%–4.5%) and high Fe/Mg ratios. Small variations in REE distribution are observed inside the Lomland dyke [La/Yb=12; LREE ca. 150–180 (chondrite-normalized values), neutral to slightly positive Eu anomaly]. Part of the Vettaland dyke is severely depleted in Zr, Rb and REE and shows a positive Eu anomaly. All rocks are depleted in U and Th, and show very low Cr and Ni contents, as well as high Sr contents (400–600 ppm). Variation within the Lomland dyke is satisfactorily ex-plained through subtraction of an apatite-bearing noritic cumulate. The role of apatite is predominant in controlling the REE behaviour. Eutectic partial melting of a Fe-rich noritic cumulate (containing apatite) under low pH2O conditions is suggested for the Vettaland dyke by REE modelling. Sr isotopic initial ratios are different in Lomland (ca. 0.708) and Vettaland (ca. 0.706) dykes. These values preclude direct derivation from the mantle as well as comagmatic relationship with anorthosite. Contamination of a mantle derived magma by deep crustal material or anatexis of the LIL depleted lower crust is compatible with the low U and Th contents, and with the Pb isotopic ratios. The Vettaland partial melting process is extended to all monzonoritic types, the necessary heat being provided by the high temperature anorthositic crystal mush.