Development of an intertidal foraminifera training set for the North Sea and an assessment of its application for Holocene sea-level reconstructions

Abstract

Regional datasets of the vertical distribution of intertidal foraminifera are useful to reconstruct Holocene sea-level changes from fossil foraminifera in estuaries and salt marshes. In this paper, we present a new foraminiferal dataset from the Ythan Estuary (Scotland) and combine it with data from eight other coastal sites from England, Denmark and Germany to produce a regional modern training set for the North Sea. We recognise a correlation between foraminifera and tidal elevation which makes the foraminifera suitable as sea-level indicators. We subdivide the data into subregional training sets and develop WA and WAPLS transfer functions. Applying a variety of statistical methods, including detrended canonical analysis, cross-validation by bootstrapping and leave-one-site-out, and the modern analogue technique, we establish the most appropriate transfer function from which to reconstruct early Holocene sea-level changes in a sediment core from the western North Sea coast. Results show that the subregional England/Scotland training set provides the most appropriate sea-level reconstructions, with decimetre-scale uncertainties. The techniques we use in this study, that consider both the modern and fossil assemblages to determine the best training set and transfer function, are suggested as a template for the development of regional transfer functions based on foraminifera and other intertidal microfossils.