Oxygen isotope data for four deep-dwelling planktonic foraminifera species collected in the subtropical NE Atlantic

Stable oxygen isotopes (δ18O) of planktonic foraminifera are one of the most used tools to reconstruct environmental conditions of the water column. Since different species live and calcify at different depths in the water column, the δ18O of sedimentary foraminifera reflects to a large degree the vertical habitat and interspecies δ18O differences and can thus potentially provide information on the vertical structure of the water column. However, to fully unlock the potential of foraminifera as recorders of past surface water properties, it is necessary to understand how and under what conditions the environmental signal is incorporated into the calcite shells of individual species. Deep-dwelling species play a particularly important role in this context since their calcification depth reaches below the surface mixed layer. Here we report δ18O measurements made on four deep-dwelling Globorotalia species collected with stratified plankton tows in the eastern North Atlantic. Size and crust effects on the δ18O signal were evaluated showing that a larger size increases the δ18O of G. inflata and G. hirsuta, and a crust effect is reflected in a higher δ18O signal in G. truncatulinoides. The great majority of the δ18O values can be explained without invoking disequilibrium calcification. When interpreted in this way the data imply depth-integrated calcification with progressive addition of calcite with depth to about 300 m for G. inflata and to about 500 m for G. hirsuta. In G. scitula, despite a strong subsurface maximum in abundance, the vertical δ18O profile is flat and appears dominated by a surface layer signal. In G. truncatulinoides, the δ18O profile follows equilibrium for each depth, implying a constant habitat during growth at each depth layer. The δ18O values are more consistent with the predictions of the Shackleton (1974) palaeotemperature equation, except in G. scitula which shows values more consistent with the Kim and O'Neil (1997) prediction. In all cases, we observe a difference between the level where most of the specimens were present and the depth where most of their shell appears to calcify.