Turbidites and iron formations, Beardmore‐Geraldton, Ontario: application of a combined ramp/fan model to Archaean clastic and chemical sedimentation

Abstract

ABSTRACTSandy turbidites, grain flows, conglomeratic mass‐flows and oxide‐facies iron formation are present in the late Archaean Beardmore‐Geraldton terrain, a metasedimentary belt which extends for at least 80 km in an E‐W direction. The marine portion of this basin contains four lithofacies associations (LA): (1) Thinbedded, iron formation‐clastic sediment association. This association represents a continuum of deposit types containing iron formation; subtypes are defined on the basis of bedding attributes and the proportion of iron formation to sand/silt. (2) Thin‐bedded, turbidite‐dominated association. These sediments consist mostly of silt/sand beds which either show no vertical trends, or thin and fine upwards over a few metres. (3) Medium‐bedded, turbidite‐dominated association. Most of these sediments are medium to coarsegrained, vertically unstructured sand sequences with occasional structured intervals. (4) Thick‐bedded association. This is dominated by poorly graded sands up to 7–8 m in thickness. Sand beds are characterised by a thin basal zone of coarse sand and pebbles, a large central interval containing a mixture of medium and coarse sand, and a thin upper zone of fine sand/silt.The overall depositional system was initiated by transport of sediment by braided streams to the strand area where it accumulated in distributary mouth bars. We infer a nearshore break in slope, locally with large channels (LA 4) extending from close to the strand line across deltaic surfaces to the deeper portions of submarine fans (structured portions of LA 3). However, many deltaic surfaces probably were not tapped by major channels, but merged downslope into a submarine ramp. Sediment was transported across the ramp by slump events and sheet‐like grain flows (unstructured portions of LA 3). Iron formation and LA 2 sediments probably accumulated both in upper‐mid ramp areas with low sediment delivery rates, and distal to fan‐ramp successions.As major streams on the braid plain changed position, associated submarine channels and slump‐fed ramp deposits also would have shifted laterally. This produced overlap of different facies associations in both the fan and ramp environments, which may explain why observed vertical trends in bedding are limited to several metres. We suggest that on narrow, active Archaean cratonic margins, additional non‐regular processes such as variations in sediment supply related to periods of heightened pyroclastic activity, and seismic activity associated with the arc, also contributed to the lack of vertically structured sequences. Turbidite sequences in such environments in general may contain important contributions from both submarine ramp and submarine fan sedimentation.