A foraminifera-based transfer function as a tool for sea-level reconstructions in the southern Bay of Biscay

A foraminifera-based transfer function has been developed using a modern dataset of 46 samples and 23 species obtained from four Basque marshes in Northern Spain. The relationship between observed and foraminifera-predicted elevation illustrated the strong performance of the transfer function (r(jack)(2) = 0.87). Indeed, these results indicate that precise reconstructions of former sea levels are possible (RMSEPjack = 0.27 m). The precision of the transfer function is comparable to other foraminifera-based transfer functions front the northern Atlantic Ocean. This is even more important considering that sampling procedures did not involve taking samples along transects that reflect the elevational gradient of concern. Instead, sampling was performed randomly to avoid possible correlation between the different variables controlling the foraminiferal distribution (e.g., elevation and Salinity). This transfer function has been constructed including agglutinated and calcareous foraminifera with the aim of increasing the range of sedimentary environments that can yield relative sea-level (RSL) data, and providing improved quantitative reconstruct ions with defined error terms in a region (Bay of Biscay) where highest marsh surfaces are Under-represented in late Holocene sedimentary sequences. (C) 2008 Elsevier Masson SAS. All rights reserved.