Using foraminiferal test-size distribution and other methods to recognise Quaternary bathyal turbidites and taphonomically-modified faunas

Abstract

Abstract The aim of this study was to develop a simple methodology using foraminifera to distinguish between sediment that has been displaced downslope (as in a turbidity current) and in-situ hemipelagic sediment. These same methods will also be useful to identify seafloor foraminiferal faunas that have suffered considerable taphonomic modification (e.g., through dissolution, winnowing, mixing), which render them unsuitable as modern analogues for paleoenvironmental reconstructions. To help develop this methodology we used Late Quaternary samples from cores containing identified turbiditic sand-mud couplets that had been deposited at lower bathyal to upper abyssal depths (1400–3000 m) off the earthquake-prone east coast of New Zealand. Comparison of the proportions of the total planktic and benthic foraminiferal tests in each of three size categories (63–125 μm, 125–250 μm, >250 μm) with those in unmodified, in-situ modern bathyal faunas from off New Zealand showed an increased proportion of larger tests in the lower sandy turbidite layers and a slight increase of smaller tests in the upper muddy layers presumably as a result of winnowing processes during deposition. A regression formula was used to estimate depositional depths based on the relative abundance of planktics in the Quaternary foraminiferal faunas (>63 μm). In all but one of our core samples, these depth estimates were considerably shallower than the cores' present depth and suggested that most sediment (both sand and mud) had been transported downslope. Modern Analogue Technique (MAT) was used to estimate depths of deposition of the core samples based on a modern benthic foraminiferal dataset (>63 μm) from around New Zealand. There was no great difference between estimates based on species % and on genus %. Some MAT depth estimates were inconsistent with those from the other foraminiferal and sedimentological proxy methods that suggested turbiditic displacement. Because of the inconsistent results and the much greater work required to acquire the species census data we exclude the MAT method from our suggested methodology. Genera were lumped into groups with similar bathymetric range and the relative abundances of these groups in each core sample were used to infer sediment provenance and mixing during downslope transport. We recommend a methodology using three elements: Test size distribution, planktic percentage, and relative abundance of bathymetric groups of benthic genera. The first two require the least amount of work and can reliably identify the majority of displaced and taphonomically-modified faunas. The third requires some taxonomic skills to identify key genera but gives more information on the provenance depth and amount of mixing in displaced turbiditic sediment. Although developed using New Zealand modern and Late Quaternary foraminiferal data, we contend that these methods should be applicable for use throughout most of the Neogene in other temperate and subtropical parts of the world's oceans. Before use elsewhere, studies in the local region may be needed to test, recalibrate and tweak the modern analogue data on local test-size distribution and planktic percentage. Our study also suggests that these methods can be useful to identify seafloor foraminiferal faunas that have suffered considerable taphonomic modification (e.g., through dissolution, winnowing, mixing), which render them unsuitable as modern analogues for paleoenvironmental reconstructions.