Holocene relative sea-level changes in northwest Ireland: An empirical test for glacial isostatic adjustment models

Abstract

The late-Quaternary relative sea-level (RSL) history of Ireland is complex, positioned at the margins of the former British-Irish Ice Sheet, and subject to the influence of ice unloading and forebulge collapse. Geophysical models of post-glacial isostatic adjustment (GIA) provide estimates of the pattern of RSL change since deglaciation which may be tested and validated with empirical data from proxy records. For the region of northwest Ireland, there is a paucity of high-quality RSL data and, therefore, equivocal evidence to support the GIA models that predict a mid to Late-Holocene RSL highstand of between +0.5 and +2 m above present. This study aims to investigate this model-data discrepancy by reconstructing RSL change from a near continuous salt-marsh sequence at Bracky Bridge, Donegal, spanning the last ca. 2500 years. We develop a transfer function model to reconstruct the vertical position of sea level using a regional diatom training set to quantify the indicative meaning and predict the palaeomarsh elevation of the core samples. A chronology is provided by a combination of 14C and 210Pb data, with sample specific ages derived from an age-depth model using a Bayesian framework. Our reconstruction shows ca. 2 m of relative sea-level rise in the past 2500 years. This is not compatible with some previously published sea-level index points from the region, which we re-interpret as freshwater/terrestrial limiting data. These results do not provide any evidence to support a Mid-Holocene RSL highstand above present sea level. Whilst none of the available GIA models replicate the timing and magnitude of the Late-Holocene RSL rise in our reconstruction, those which incorporate a thick and extensive British-Irish Sea Ice Sheet provide the best fit.