Mineralogy and geochemistry of pattern formation in zebra rock from the East Kimberley, Australia

Abstract

Rhythmic patterns are widespread in geological materials. A particularly striking, macroscale example is zebra rock from the East Kimberley region of northwestern Australia. The rock is famous for its distinctive, rhythmically ordered, iron-oxide pattern, which transforms an Ediacaran-aged siltstone into an attractive semi-precious gemstone. Several different formation mechanisms of this pattern have been proposed in previous studies, with the two most prominent being redoximorphic banding in acid-sulfate soils and Liesegang banding in an acidic hydrothermal system. Using a combination of mineralogy, geochemistry and geological context, this study attempts to confirm both the occurrence and relative timing of acid-sulfate fluid rock interactions in zebra rock and seeks to determine whether pedogenic processes or hydrothermal alteration can better explain the origin of these patterns. We present the first evidence that the iron-oxide banding developed simultaneously with a period of aluminosilicate dissolution, clay precipitation, and fluid flow, consistent with the infiltration of an acidic fluid. This conclusion was evidenced through the hexagonal-platelet morphology of the hematite pigment, kaolinite-hematite textural relationships, and a paleoflow direction preserved in the asymmetric intensity of the hematite concentration. However, while the mineralogy of zebra rock strongly suggests interactions with acid-sulfates, the origin and temperature of the fluid could not be conclusively determined. Supporting a hydrothermal origin, a thorough analysis of zebra rock mineralogy revealed a mineral assemblage consistent with advanced argillic hydrothermal alteration, wherein pyrophyillite, kaolinite, and dickite indicate minimum palaeotemperatures upwards of 120 °C and the presence of alunite and a svanbergite-woodhouseite solid solution suggests oxidising, acidic (pH <5) conditions. The low Rb/Sr ratio and relative immobility of rare earth elements in most zebra rock deposits are also consistent with an acidic hydrothermal origin. Further support was also observed in the mineralogical trends between examined outcrops, grading from alunite-type to kaolinite/dickite-type facies in a south-west direction. But despite this evidence, the acid-sulfate soil hypothesis could not be refuted and was itself supported by the large number of pyrite dissolution voids both underlying and within the patterned layer. Furthermore, the consistent compaction of reduction spheroids, dissolution voids, and pseudomorphic inclusions within the light banding of zebra rock are in agreement with near-surface supergene weathering, ruling out hypogene hydrothermal alteration. A mechanism of pattern formation is proposed whereby zebra rock banding is formed by the Liesegang phenomenon, driven by the oxidation of Fe2+ ions during the infiltration of an Fe2+-rich, acid-sulfate fluid into oxidising host sediments.