A robust understanding of coccolithophorid ecology is of great potential value since: (1) They constitute a significant component of the modern phytoplankton. (2) The exceptional fossil record of individual species gives them great potential as palaeoecological proxies (e.g. Molfino & McIntyre 1990, Chepstow-Lusty et al. 1992, Eshet et al. 1996). (3) Total coccolithophorid production is readily determinable from carbonate sediment accumulation rates and constitutes the most accessible single geological proxy for total phytoplankton productivity (Brummer & van Eijden 1992 , Siesser 1995). In addition, lipid biomarkers provide an independent record of coccolithophorid production (Conte et al. 1994). (4) Pelagic carbonate sedimentation has a major impact on the global carbon dioxide cycle (Milliman 1993, Wollast 1994).
General aspects of coccolithophorid biogeography and habitat are well-know from taxonomic surveys of the plankton and of bottom sediments (e.g. Kleijne 1993, Brand 1994, Roth 1994, Winter et al. 1994, Young 1994, Samtleben et al. 1995). Individual species are cosmopolitan but with more or less limited latitudinal distributions. Thus, 3 or 4 broad coccolithophorid zones are recognised (McIntyre & Be 1967, Okada & Honjo 1973, Winter et al. 1994). These latitudinal zones must in large part reflect temperature tolerance but even at this level it is apparent that other factors are of major significance. For instance Coccolithus pelagicus which largely defines the high latitude zone occasionally occurs at much lower latitudes (e.g. Pujos 1992, Giraudeau 1993, Cachao subm.), and was widespread in the warmer conditions of the Late Pliocene. Similarly the presence of floral zones associated with equatorial upwelling and the sub-topical oligotrophic gyres suggest control by nutrients/trophic level rather than temperature alone.
We have only limited understanding of the ecology of a few individual species, notably Emiliania huxleyi. This reflects, in part, a paucity of culture studies and so, of direct information on the autecology of species. It also reflects a shortage of suitable studies on natural populations. Much available data deal with relative numerical abundances in coccolithophorid assemblages, which is likely biased. Little work has been done on abundances relative to total phytoplankton or on absolute abundances of single species relative to ecological parameters in modern communities. In contrast, extensive geological studies have used coccolithophorid assemblage changes as indicators of palaeoceanographic conditions (e.g. Haq 1980, Aubry 1992, Jordan et al. 1996). However, such studies have not been outstandingly successful because they lack the biological models necessary to interpret, non-linear, responses to palaeoceanographic or climatic conditions.
Similarly, there is no clear understanding of the controls on the coccolithophorid component of total phytoplankton primary production, but rather a set of loosely defined and partially conflicting hypotheses. Hypothesis 1: Coccolithophorids do not fill any particular ecological niche but rather have evolved a disparate range of ecological strategies and hence total coccolith production is a random function of the evolutionary success of individual species. Hypothesis 2: Coccolithophorids are an oligotrophic adapted group, K-selected species specialised for low nutrient, high stability conditions. Hence they may provide a good index of open ocean primary productivity (Brummer & van Eijden 1992). Hypothesis 3: Coccolithophorids, or at least the dominant species in terms of total coccolith production are essentially a eutrophic adapted group but are outcompeted by diatoms until silica is depleted. Hence total coccolith production is a combined function of macronutrient and silica distribution (E.g. Egge 199x). Hypothesis 4: Coccolithophorids occupy a distinctive ecological niche because they calcify - and so replace diatoms when CO2 becomes limiting (e.g. REF).
Our study will provide a multiple approach analysis of the ecology of the key species. This will both allow these important species to be used as individual proxies and test the above proxies in order to produce an understanding of the ecology of the coccolithophorids as a group, allowing critical interpretation of the bulk coccolith accumulation record and modelling of the role of coccolithophorids within the global carbon cycle.
Ecological interpretations based on any one type of observation are, almost inevitably, ambiguous. Consequently our approach is to combine observations from a range of methodologies each with different strengths and weaknesses so that any interpretation from one type of observation can be cross-checked with data from other types of observation. In addition the combination of laboratory and field observations allows explicit testing of questions, such as: do species consistently occupy the physico-chemical conditions to which they are optimally adapted, or do they occupy different conditions as a result of competitive displacement? The range of observations types to be used are:
1. Study of physiological adaptation in laboratory culture experiments. In particular light-temperature cross-gradient cabinets will be used to determine optimal growth conditions in culture across a range of physico-chemical conditions. (NHM, ETHZ).
2. Cruisework study of development of coccolithophorid populations in relation to hydrographic conditions and phytoplankton succession (FdA-VUA, CSIC, MNHN-UL).
3. Filter sample and sediment trap sample studies of seasonal succession and depth preference (ETHZ, FdA-VUA, MNHN-UL).
4. Holocene sediment sample study in order to map out global biogeography relative to large scale oceanographic parameters (especially nutrient distribution, and temperature). (FdA-VUA, ETHZ, MNHN-UL).
5. Geological sample study in order to examine distribution of individual taxa relative to known parameters of oceanographic change. This work will be carried out on DSDP/ODP sites for which there are well constrained background studies, including detailed age models. The work will be based on use of quantitative techniques to allow calculation of accumulation rates of taxa (i.e. liths per unit area of sea-bed per unit time) - Beaufort (1991), Su (1996), Lototoskaya (in prep.). It will be combined with morphometric study of morphological evolution. (NHM, MNHN-UL, ETHZ).
6. Calculations of gene flow. Microsatellites generated from phylogenetic analyses of closely related morphotypes of Gephyrocapsa can be used to estimate gene flow and hence niche size. Populations with high gene flow are generally more cosmopolitan than those with low gene flow and hence have a broader niche size. (AWI)
7. Study of biolipids. Lipid analyses will be carried out on selected samples from the sediment trap and geological data sets to compare the lipid biochemical record of coccolithophorids with conventional assemblage data (NIOZ).
The distribution analyses (2-5) will be primarily based on quantitative analysis of absolute abundances (e.g. specimens per litre of seawater) this allows detailed investigation of the distribution of individual species from large numbers of samples. Morphometric work will be used to characterise populations in terms of morphotype distribution and to allow calibration of coccolith volume estimates so that species-specific carbonate fluxes can be calculated.
The selected keystone species do not include deep photic or oligotrophic specialised forms and so are not representative of the full diversity of coccolithophorid ecological adaptations. They do, however, include most of the larger abundant taxa and dominate carbonate fluxes in most modern environments (our data from sediment trap studies). Their direct ancestors similarly dominate most Cenozoic nannofossil assemblages. Hence biological calibrations based on these taxa are directly applicable to modelling of the role of coccolithophorids in the carbon cycle.
Our basic objectives for each species are to:
KEY SYNTHETIC OBJECTIVES ARE TO:
This page is based on the CODENET proposal, as accepted for funding by TMR, June 1997