Pisolitic laterite geochemistry in the golden grove massive sulphide district, Western Australia

Abstract

Abstract The two known massive Cu-Zn sulphide deposits in the steeply dipping strata of the Golden Grove District of Western Australia — the Gossan Hill Deposit and Scuddles Prospect — both have a multi-element secondary geochemical dispersion halo in the pisolitic layer of laterite of probable Tertiary age. The Gossan Hill geochemical anomaly is about 1.5 × 2 km and has a Bi-Sn-Mo-In-Sb-As-Cu-Au association. Mean compositions of pisolitic laterite samples range from 4 to 18 times background, with maxima for Bi, Sn and Mo reaching 90 to 100 times background. Relief of Gossan Hill, which rises some 80 m above the laterite plain, would have favoured dispersion during lateritization. Nevertheless at Scuddles, where the laterite surface has only a gentle slope across strike, an anomaly in pisolitic laterite measures 1 to 1.5 by 1.5 to 2.0 km. The anomalous association, Bi-Sn-Mo-Sb-As, is similar to that at Gossan Hill; the mean contents in pisolitic laterite samples from the Scuddles anomaly range from 2.3 to 8 times background, with maxima for Bi being 12 times background. At Gossan Hill, individual pisolites and the concretionary skins contain gossan fragments and cassiterite, indicating mechanical dispersion during interitization. However, anomalous Cu, Zn and As in goethites of the concretionary skins of pisolites and nodules from the centre of the anomaly indicate that there has also been hydromorphic dispersion. The geochemical characteristics of pisolitic laterite samples have been represented by empirically derived additive indices based on up to eleven chalcophile elements. One index is suggested as a method to increase anomaly size and simplify interpretation; a more specific index based on Bi, Sn and Se is suggested for defining more closely the actual centres of mineralization. For routine exploration of lateritic terrain where the dips of the host sequence are steep, 320-m spacing of samples on a triangular grid appears suitable for locating anomalies and generally defining their source. However, for reconnaissance exploration, the density of sampling could be significantly reduced for example to 1 km spacing.