Simulating our ability to accurately detect abrupt changes in assemblage-based paleoenvironmental proxies

Abstract

Resolving abrupt environmental changes in sedimentary records is critical to understanding environmental perturbations relevant on human timescales. The paleontological assemblage mixer (paleoAM) framework developed here simulates sedimentary records to measure the preservation potential of abrupt changes in assemblage-based faunal proxies while varying environmental background conditions, excursion magnitudes and durations, bioturbation, sedimentation rates, and sampling completeness. Using a record of fossil benthic foraminifera, we apply paleoAM to quantitatively determine how distinct from background conditions and how enduring an assemblage change must be to be accurately detected. At the high sedimentation rates of the case study, century-long low-oxygen events frequently have a high probability of being sampled and accurately detected, quantitatively estimated with simulations across varying sedimentation rates, and the assemblage-based proxy is robust to bioturbation. Even brief events are frequently distinguishable from background conditions, particularly for extreme events, although event magnitudes are underestimated, consistent with the proportional mixing of background and event assemblages. Event magnitudes are never overestimated, indicating that single-sample assemblage excursions observed in empirical records are true shifts. Accurate detection predictably declines when bioturbation is permitted, sedimentation rates are lower, samples are stratigraphically thicker, or sample spacing is greater. Further, paleoAM can estimate the sampling strategy needed to detect abrupt events given a small set of representative "pilot" samples. The paleoAM framework is adaptable to any abundance distribution and event scale to create informed sampling plans and to assess the ability of sedimentary records to resolve target events given robust estimates of sedimentation and potential mixing rates.