The surface ocean productivity response of deeper water benthic foraminifera in the Atlantic Ocean

We test the relationship of deep sea benthic foraminiferal assemblage composition to the surface ocean productivity gradient in the low latitude Atlantic Ocean using 81 surface sediment samples from a water depth range between 2800 and 3500 m. The samples are selected so that the surface ocean productivity gradient, controlling the flux of organic carbon to the seabed, will be the most important environmental variable. The first two principal components of the assemblage data account for 73% of data variance and are clearly linked to the productivity gradient across the Atlantic. These components, show that under higher productivity the assemblages contain a higher abundance of Uvigerina peregrina, Melonis barleeanum, Globobulimina spp. and other taxa with probable infaunal microhabitats. Alabaminella weddellensis, a species linked to episodic phytoplankton debris falls, is also important in these assemblages. As productivity decreases there is a regular shift in assemblage composition so that low productivity assemblages are dominated by Globocassidulina subglobosa and several Cassidulina species along with Epistominella exigua. We hypothesize that these taxa are epifaunal to very shallow infaunal since nearly all organic carbon oxidation occurs near the sediment-water interface in low productivity settings. Discriminant function analysis of the foraminiferal assemblages, with groups selected on the basis of surface ocean productivity, shows clear separation among five productivity levels we used. This analysis demonstrates that productivity variations have a strong influence on assemblage composition. Finally, we used two groups of samples from the Rio-Grande Rise representing water depths from 2007 to 2340 m and 2739 to 3454 m to test for effects produced by changing water depth. All these samples are from a low productivity region and represent nearly identical environmental conditions. Although the low productivity nature of all the Rio-Grande Rise samples is obvious, there are assemblage differences between our depth groups. We cannot account for the assemblage differences with changes in organic carbon flux, dissolution effects or other physical/chemical properties of the ocean. Thus there are as yet unidentified factors related to water depth which cause some assemblage variation in the low productivity setting we investigated.