FORMATION AND DEFORMATION OF THE MICROFOSSIL RECORD: PLANKTON, SEDIMENT TRAPS AND SURFACE SEDIMENTS

Schematic diagram illustrating the various mechanisms that distort the sedimentary imprint of the planktonic patterns of fossilizable microplankters (see text for detailed explanation).

The fossilizable remains of marine planktonic organisms (chiefly coccolithophorids, silicoflagellates, diatoms, dinoflagellates, radiolarians, foraminifers, pteropods, ostracods) have been used intensively for paleoecologic reconstructions for over 50 years. Taxonomic studies reached their climax in the mid 70's to early 80's, thereafter emphasis gradually switching to isotopic analyses. The wealth of information amassed throughout the last decades allowed producing detailed interpretations of climatic changes covering time-scales from annual seasons to millions of years. However, although the value of these results is beyond doubt, most of the surveys ignored or minimized the effects of various phenomena that have a direct and profound impact on the mechanisms of sediment formation, mechanisms that only recently we are starting to examine in detail, thanks largely to the advent of new technologies unavailable a decade ago. Indeed, paleoecologic interpretations of pelagic sediments based on floral and faunal studies rely largely on the assumption that bottom thanatocoenoses are the product of the rather uncomplicated and straightforward vertical sinking of the overlying planktonic assemblages. Because near-surface communities are the ones most influenced by, and most tightly coupled with climatic changes, and since these near-surface layers host highest planktonic densities (thus, in turn, yielding highest sedimentary outputs), it is usually explicitly or implicitly assumed that the environmental signal buried in the sedimentary record represents an adequate proxy of variations in the epipelagic realm and, hence, in gobal or regional climate. Hints suggesting that sedimented assemblages differ significantly from those present in the plankton were furnished by many surveys which drew direct comparisons between the two, yet they generally were restricted in scope and lacked crucial data necessary to pinpoint breakpoints in the mechanisms of transfer of the planktonic biocoenoses to the bottom. Some important details of these mehanisms were furnished in the last years by sediment trap materials. Samples collected by time series sediment traps deployed at mid depths in various oceanic regions represent an intermediate step in the formation of the microfossil record, and therefore allow interpreting some of the constraints potentially responsible for the plankton-sediments uncouplings observed.

Differences in species makeups and proportions in the near-surface plankton and in the surface sediments can arise as a consequence of post-mortem processes, of processes associated with traits of the living organisms, or a combination of the two. Post-mortem distorting mechanisms include fragmentation (I in figure) and bulk and selective dissolution en route to the sea-floor (B) and on the bottom (C), vertical reworking of sediments (D), and winnowing and lateral advection of displaced sediments (H). Among the processes associated with traits of the living organisms of special importance are differential seasonal dynamics (G), differences in species-specific productivity (F), equatorward subsurface transport (A), and integration of surface (environment-sensitive) and deep (environment-insensitive) species in the sedimentary record (E).

Below 10% of the calcium carbonate and the biogenic silica produced in the water-column is preserved in the fossil record; furthermore, much of this accumulation is in the form of unidentifiable fragments produced as a result of grazing by larger organisms (I), useless for paleoecologic purposes. In some cases (e.g., coccolithophorids, phaeodarians) most or all the skeletons disappear in the process of sinking (B). Fragmentation due to grazing further enhances dissolution. An especially disturbing circumstance is that fragmentation and dissolution are not evenly spread over the entire planktonic assemblage because more fragile species are affected more than the robust skeletons. As a result, the dominant components of the plankton can be totally wiped out off the surface of the ocean bottom. This type of selective destruction can have a profound impact on sediment-derived paleotemperatures because in several microplanktonic groups coldwater taxa have more dissolution-resistant tests than the warmwater ones. The selective preservation of these more robust species in mixed sediments can thus enhance their relative importance over the (dissolved) "warmer" skeletons.

Reworking of surficial sediments by benthic organisms that burrow in the upper 5 to over 40 centimeters mixing older (deeper) material with younger (shallower) deposits is a widespread phenomenon in all ocean basins (D). Sediment accumulation rates over much of the World Ocean rarely exceed 1-3 cm per 1,000 years, which implies that deposits retrieved in core top samples can be as old as 30,000-40,000 years. Since reported world-wide climatic changes occurred on time-scales comparable to those at which reworking operates, models based on sedimentary materials can lump traits of Recent patterns with those of distribution ranges characteristic of past oceanographic settings. Only a few basins where restricted bottom water circulation and high input of organic matter make benthic conditions anoxic (e.g., the Santa Barbara and Santa Monica basins off California) can host finely layered sedimentary sequences undisturbed by this process.

Strong subsurface currents can significantly displace shells settling to the sea floor so that sedimentation occurs outside of the "home range" area in the plankton (Fig. 1A). In the eastern Atlantic, for example, Antarctic diatoms occur as far north as 2°S in trap samples deployed at 700 m . The Subarctic radiolarian Lithomitra arachnea is a dominant component of California Borderland sediments, yet this species is totally absent from the overlying 0-100 m plankton throughout the year. Lateral advection of near-surface-dwelling plankters and their subsequent sedimentation in areas outside of the living range of the species (A) can be strongly enhanced by submergence of the living organism as a means of counteracting expatriation into hostile environments. Thus, when a coldwater organism is carried by currents into a warmer area, it can sink in the water column in search of colder strata; subsequent displacements in the same direction will force it to continue descending until it perishes and sediments when its tolerance threshold is exceeded. Bipolar distributions connected at depth across the equator indicate that this behavioral pattern is rather common in the ocean. As a result of this phenomenon, many coldwater species which in the plankton do not extend beyond their characteristic water masses, are present in the sediments throughout the entire subtropical, tropical and equatorial belts (e.g., the foraminifers Globoquadrina (=Neogloboquadrina) pachyderma, Globigerina quinqueloba, G. bulloides, Globorotalia scitula). Input of warmwater remains in colder-water areas has also been reported. For example, the presence of Atlantic transitional and subtropical radiolarians in the Scotia and Weddell seas at depths below 300-400 m is most probably the result of lateral advection with the Warm Deep Water.

Lateral advection can also occur after sedimentation, as a rersult of bottom currents that resuspend and displace newly deposited material thus effectively precluding sediment accumulation over vast areas of the sea-floor (H). Antarctic Bottom Waters, for example, are responsible for large hiatuses in the sedimentary sequences of several sites around the Antarctic where sediments from the last 3 to 6 million years are absent altogether. This mechanism not only eliminates the record from its original area, but it also carries the signal elsewhere thus "contaminating" other regions with fossil remains unrelated with the corresponding upper-layer assemblages. The diatom Nitzschia kerguelensis, entrained in Antarctic Bottom Waters, has been recorded as far north as 8°S in the Indian Ocean; this species is light enough to be carried over considerable distances, yet well silicified as to resist erosion, dissolution and fragmentation. Coscinodiscus lentiginosus can be displaced in the South Pacific, and others described the northward displacement of Eucampia antarctica in the Argentine Basin.

Although seasonality as such is only preserved in anoxic, varved sediments, seasonal differences in the production, sedimentation and grazing of plankton can engender sedimentary assemblages strongly biased toward the season when production and output are highest, and destruction by grazing is lowest (G, I). In the Weddell and Scotia Seas over 80-90% of the overall yearly radiolarian production is circumscribed to a short (1-2 months) period of ice-free waters. The environmental signal of diatom thanatocoenoces from British Columbian fjords is severely distorted by seasonal variations in the output of frustules and different modes of grazing by the dominant zooplankton. It is conceivable that the fast reproduction rates of most autotrophic plankters (cocolithophorids, diatoms, silicoflagellates), as compared with those of the heterotrophic ones, make their sedimentary records more prone to bias due to seasonal differences in production rates because their abundances in response to short-term environmental shifts can yield faster and larger fluctuations. Thus, while phytoplankters are more sensitive to ambient changes in the uppermost layers of the ocean, zooplanktonic water-column assemblages smooth out extreme conditions thus yielding an integrated record less affected by the short periods which deviate most from the long-term average.

Differences in species-specific productivity can yield sedimentary assemblages where organisms with shorter life spans and higher turnover rates are overrepresented with respect to those that live longer and reproduce more slowly (F). It has been suggested, for example, that smaller foraminifers have shorter life spans than larger ones, which could in principle bias shell output in favor of the former.

In bottom deposits, the record of the near-surface environmentally meaningful taxa is mixed with remains engendered at meso- and bathypelagic depths, where changes in association with climatic shifts are much more subtle (E). Although planktonic densities at these deep layers are low, deep species usually inhabit rather large vertical distances; thus, integration of low densities over large depth-intervals can yield high proportions of meso- and bathypelagic species in sedimentary sequences. For example, the output of radiolarians living below 300 m can exceed by over 4 times that of the species living between 0 and 100 m.

Although our present state of knowledge does not allow unequivocal identification of the various factors responsible for the lack of similarity between live planktonic assemblages and their sedimentary imprints, it is becoming increasingly obvious that significant modifications are neither restricted to, nor probably dominated by the post-depositional diagenetic processes advocated commonly, but occur in the upper hundreds of meters of the water-column. Acknowledgement of these distortions is a sine qua non condition for the sound utilization of the fossil record for paleoecologic and paleobiogeographic surveys, while their investigation should benefit from an interaction between geological, oceanographic and biological evidences.

Back to Main Home Page