The origin of modern oceanic foraminiferal faunas and Neogene climate change

Planktonic and benthic foraminifera are the most significant providers of information on the state of surface and deep oceans in the past. Many foraminiferal proxies rely on the knowledge of ecological preferences of individual species and the assumption that these remained similar through time. Consequently, the applicability of such proxies is limited in time by the extent of the modern fauna. By analysing extensive datasets of species occurrences, we show that the modern oceanic foraminifer fauna originated during the Neogene. This occurred during two distinct diversification pulses: one in the Middle Miocene (17–14 Ma) and the second at the Miocene/Pliocene transition (7–4 Ma). The first diversification coincides with the time of a major change in the frequency of the dominant climate cycles during the Miocene Climatic Optimum. The environmental driver of the second diversification could be related to an increased provincialism induced by the closure of the Panama Seaway, but the exact link is not clear, particularly for the plankton. Surprisingly, major changes of ocean circulation due to the growth of Antarctic ice-sheet and closure of low-latitude seaways appear to have caused mainly extinctions. Given the age of the latest diversification and extinction pulses that shaped the modern foraminiferal fauna, we conclude that calibrated proxies based on assemblage properties should not be interpreted quantitatively in sediments older than the late Pliocene. Much of what we know about the Cenozoic oceans comes from the study of fossil foraminifera. These microscopic protozoans occur in nearly all marine environments – from the bottom of the deepest trenches (Todo et al. 2005) to brine channels in Antarctic sea-ice (Dieckmann et al. 1991) – and their ornate shells are readily preserved in sediments. Foraminiferal species are sensitive to a range of environmental parameters and the chemical and isotopic composition of their calcite skeleton records the properties of ambient seawater and the nature of metabolic and kinetic processes that take place during calcification. This remarkable capacity to reflect the state of their habitat, coupled with the abundance of their fossils in marine cores, made foraminifera the main tool for reconstructions of past oceans and climate. Benthic foraminifera convey information about conditions prevailing on the ocean floor. Early research on benthic foraminifera was qualitative, aimed at assessing the faunal inventory, and obtaining a general understanding of past environmental conditions. Recent studies focused on the development of quantitative tools, in particular for the reconstruction of organic matter flux to the sea floor (Herguera & Berger 1991; de Rijk et al. 2000), oxygen content of near-bottom waters (Kaiho 1994), and intensity of bottom currents (Schonfeld 2002). Planktonic foraminiferal assemblages have long been used to estimate sea-surface temperature (Imbrie & Kipp 1971; Kucera et al. 2005), upwelling intensity (Thiede 1975; Conan et al. 2002) and productivity (Ivanova et al. 2003), and the modification of their assemblages on the sea floor has been used to reconstruct past bottom-water carbonate ion concentration (Conan et al. 2002). The elemental and isotopic chemistry of calcareous tests from both benthic and planktonic foraminifera forms the basis of an ever-increasing battery of proxies used to reveal the age, origin, chemical and physical properties of ocean waters (Fischer & Wefer 1999; Henderson 2002). The application of foraminifera as palaeoceanographic proxies is based on the recognition of ecological relationships in the modern ocean and their translation into the fossil realm. This method relies on a number of assumptions (e.g. Birks 1995), the most pertinent one being the stationarity through time of the individual ecological relationships. Some foraminiferal proxies are based on determination of broad functional and ecological types (benthic vs. planktonic) or preservational state (fragmentation) and their temporal applicability is therefore less of an issue. However, most quantitative proxies rely on species-specific calibrations and the interpretation of almost all chemical From: WILLIAMS, M., HAYWOOD, A. M., GREGORY, F. J. & SCHMIDT, D. N. (eds) Deep-Time Perspectives on Climate Change: Marrying the Signal from Computer Models and Biological Proxies. The Micropalaeontological Society, Special Publications. The Geological Society, London, 409–425. 1747-602X/07/$15.00 # The Micropalaeontological Society 2007. signals in foraminifera requires knowledge of the habitat and phenology of the analysed species. Because of circular evidence, the stability of the ecological preferences of taxa through time is often difficult to assess. Therefore, morphological similarity has been used as a first approximation of ecology and the fossil range of a given species has been taken to represent the duration of its specific, present-day ecological behaviour. This approximation is inevitably crude, as it is known that species’ ecology can change with little or no shift in morphology (Norris et al. 1996; Kucera & Kennett 2002), but it provides a consistent and objective means to estimate the maximum range of modern environmental calibrations. The purpose of this study is to review data on the age of extant species of planktonic and benthic foraminifera, determine the diversification history of the groups and assess its links with oceanographic events in the Cenozoic. A further aim is to produce objective guidelines for the use of foraminiferal proxies based on modern environmental calibrations in ‘deep time’. This review will only deal with oceanic foraminifera; shelf seas, brackish facies, marshes and other marginal environments will not be considered. Origin of the modern foraminiferal fauna The number of living benthic foraminiferal species is estimated at approximately 4000 (Thies 1991), and about 45 living planktonic species are commonly recognized (Hemleben et al. 1989). The number of all benthic and planktonic species since the Cambrian is estimated at about 10,000 (Vickermann 1992). This estimate would suggest that foraminiferal faunas are dominated by conservative, long-ranging taxa and that the modern fauna should have its roots in deep time. A direct assessment of the antiquity of extant foraminiferal species would require a thorough analysis of species occurrences on a global scale. However, information on the ranges of living benthic foraminifera is rather scattered. Biostratigraphic compendia typically cover a limited number of stages or a subsystem, have a regional bias and concentrate on stratigraphically important taxa. Only a few studies in the framework of the Deep Sea Drilling Project and Ocean Drilling Program go beyond that scope (see Berggren & Miller 1989; Thomas 1992; and Hayward 2001 for further discussion). Jones (1994) gives a list of stratigraphic ranges for 305 recent, mostly deep-sea benthic foraminifera and 26 planktonic species figured by Brady (1884). Less than 10% of the presumed living benthic species and only a half of the living planktonic foraminifera are covered by this compilation. Nonetheless, the dataset is methodologically and taxonomically consistent and provides the first approximation of the distribution of ages of modern oceanic foraminifera (Fig. 1). As many as 13 (4.2%) of the living benthic species listed by Jones (1994) have originated in the Cretaceous; the oldest was reported as being of Valanginian age. The number of survivors from the Palaeogene is with 58 (19.0%) still low. Clearly, the Miocene was the crucial time for the diversification of the modern benthic fauna. More than 50% of the recent benthic species included in Jones’ (1994) compilation have originated during this stage. Remarkably, almost a quarter of the modern species appear to have originated during the last 5 Ma in the Pliocene or Pleistocene. 0 20 40 60 80 100 120 140 160 180 Pleistocene Pliocene Miocene Oligocene Eocene Palaeocene Cretaceous O rig in s of r ec en t f or am in ife ra Planktonic species