Planktic foraminifera are the only extant group in which all species can be preserved as fossils in the geologic record. Their shell remains contribute to the oceanic carbonate sediment on about 70% of the ocean area which is, therefore, a source for global studies of planktic foraminifera (Belyaeva, 1976). Sediment assemblages of planktic foraminifera reveal distinct biogeographic pattern which reflects patterns in the plankton (Hutson, 1977; Bé and Tolderlund, 1971; Bé and Hutson, 1977). The relative abundances of species in the sediment indicate ecologic preferences and maximum relative abundances correlate with ecologic optima (Hecht, 1976). Consequently sediment assemblages can be used in studies of the ecology of planktic foraminifera with certain limitations and advantages of the sediment record.
The species composition in sediment assemblages reflects the relative intensity of shell production and preservation of the individual species under conditions of the environment and taphonomic effects. Environmental conditions include physical, chemical, and biological factors. In this paper I will analyse the importance of physical conditions for relative abundances of planktic foraminifera based on counts in sediment surface (Holocene) samples by Bé and Hutson (1977) and Hutson and Prell (1980) for the Indian Ocean and Kipp (1976) for the North Atlantic, which have been revised for CLIMAP (1976, 1981). Some species are primarily related to chemical and biological conditions. For these species the analysis will be incomplete. Some information on their ecologic preferences can be extracted from geographic pattern in their relative abundances.
Sinking velocities of empty shells are in the order of hundreds of meters per day leading to deposition within one to three weeks at any depth in the ocean (see data and references by Takahashi and Bé , 1985). In this time a lateral transport of the sinking tests occurs which leads to differences between living and sedimented assemblages at a given point in the ocean. Living assemblages of planktic foraminifera, however, are known to characterise water masses and are subject to the limited mixing with other water masses due to physical contrasts in the water. This applies for living and dead shells and will tend to preserve the relations of surface water masses with a characteristic assemblage of planktic foraminifera. Current directions in the intermediate and deep water often do not correspond to those in the surface waters. This effect produces complicated trajectories until a shell settles on the sea bottom and these trajectories may vary with the seasonal dynamics in the water masses. A site of deposition, therefore, collects faunas from a certain area and the combined effect of current directions, current speeds, and residence times of shells in the different water masses introduces noise to the data. This noise affects the analysis of those species more seriously that live in geographically confined areas because relations with small water masses or distinct oceanographic features may be obscured.
Selective dissolution and other effects of preservation alter the species composition of sediment assemblages. Bé and Hutson (1977), Hutson (1977), Hutson and Prell (1980), and Kipp (1976) were aware of this problem. CLIMAP Project Members (1976, 1981) eliminated problematic samples but effects of preservation on faunal assemblages are not excluded.
Bioturbation leads to mixing of assemblages. The mixed layer in the Atlantic is about 5 cm thick, sedimentation rates of oceanic carbonate are typically in the order of 1 cm/Ka (e.g. Seibold and Berger, 1993). These relations implie a long-term averaging of assemblages that were produced over some thousands of years under variable environmental conditions. Comparisons between oceanographic measurements and sediment assemblages of microfossils will, therefore, never produce ideal correlations even if such relationships exist in some species.
The above discussion may suggest serious limitations in the analysis of sediment microfossil assemblages for the assessment of ecological data. Depth-controlled plankton tows and sediment traps are probably more suitable for such analyses.
Most plankton net studies of planktic foraminifera, however, suffer from spot sampling at an arbitrary time and do not allow to assess changes in the assemblages during the bi-monthly to annual reproduction cycles of the various species at a
given location (see maps by Bé and Hutson, 1977). The seasonal dynamics is often not represented in plankton net samples. Sediment trap samples reveal substantial changes in assemblages on seasonal time scales and allow correlation of
these changes with oceanographic parameters. Few sediment traps, however, have been deployed over longer time and in fact a multi-year record for the Atlantic and Indian Ocean is presently available only from Bermuda (Deuser and Ross, 1989;
Deuser et al., 1981). I will demonstrate for some species how such data supports the interpretation of sediment assemblages and subsequently how sediment trap assemblages collected from small ocean areas demonstrate reliability of data
derived from sediment assemblages on biogeographic scales. Geologic applications, however, rely on the calibration of relations between micropaleontologic tracers and environmental conditions based on sediment samples. CLIMAP (1976, 1981) is
an example of the advantages and disadvantages of this approach. The goal of this paper is two-fold: to add to ecologic knowledge on planktic foraminifera, and to facilitate their applicability for paleoenvironmental analyses.