Strategically situated for high-latitude paleoceanographic studies in the southern Indian Ocean, the Kerguelen Plateau contains a record of changing paleoenvironmental and paleoceanographic conditions extending back to Early Cretaceous time. During the Cretaceous period, characterized by relatively warm ocean waters and high eustatic sea level, the Kerguelen Plateau was constructed in a small but expanding Indian Ocean. During Tertiary and Quaternary time, global climate cooled and ice sheets waxed and waned. Today, most of the Kerguelen Plateau lies south of the Antarctic Polar Frontal Zone (Antarctic Convergence) and beneath the main flow of the Antarctic Circumpolar Current. Leg 183's six drill sites on the plateau have yielded important information on Cretaceous paleoenvironments and on the Tertiary development and evolution of these two major physical oceanographic features, which have had and continue to have major effects on global climate and surface water circulation.
Paleobotanical and palynological data from Site 1138 on the CKP (Figs. F1, F2) provide a detailed picture of the mid-Cretaceous geological and paleovegetational history of the Kerguelen Plateau. Dark, organic-rich silty claystone with many wood fragments and fern remains, certainly of terrestrial origin (Mohr et al., this volume), directly overlies the ~100- to 101-Ma volcanic basement (Duncan, 2002). These terrestrial strata are of late Albian to earliest Cenomanian age, as dated by sporomorphs, and show that parts of the CKP were subaerial at least until late Albian time (Mohr et al., this volume). A diverse high-latitude flora, probably in a dense conifer forest with various fern taxa and early angiosperms in the undergrowth, covered the CKP. Although the vegetational character did not change over this time interval, varying percentages of several sporomorph groups seem to show possibly cyclic abundance variations, perhaps caused by Milankovitch-type events (Mohr et al., this volume). Uppermost Cenomanian to Coniacian sediments are of open marine origin and contain high-diversity dinocyst assemblages.
Late Cretaceous and Danian foraminifers recovered from Site 1138 on the CKP, a fishing vessel's dredge on the CKP, and Eltanin core E54-7 in the Labuan Basin (Figs. F1, F2) have yielded paleoceanographic information about that time interval (Quilty, this volume; Holbourn and Kuhnt, 2002). The benthic foraminiferal succession at Site 1138 records the evolution of the CKP from a subaerially exposed platform in Cenomanian time to a bathyal, pelagic environment in early Turonian (Holbourn and Kuhnt, 2002). The pelagic assemblages strongly resemble temperate, open-shelf, bathyal assemblages in the Northern Hemisphere, reflecting prevalent warm, high-latitude temperatures, open oceanic gateways, and a dynamic trans-hemispheric global circulation. The distribution of upper bathyal benthic foraminifers was strongly modulated by carbon flux and fluctuations in oxygenation, although local hydrography and taphonomic processes also played an important role (Holbourn and Kuhnt, 2002). Biofacies changes during late Cenomanian and early Turonian time relate mainly to environmental change and taphonomic bias and do not appear to be linked to true extinction and radiation events (Holbourn and Kuhnt, 2002). Cenomanian-Turonian time was warmer than the subsequent Campanian-Maastrichtian epochs, when the region was in the cool Austral Faunal Province (Quilty, this volume). Late Campanian and Maastrichtian benthic foraminifers are dominated by epifaunal species, suggesting little upwelling and a high degree of oxygenation. Late Maastrichtian foraminifers from the dredge site have a high infaunal content, consistent with significant upwelling and lower oxygenation near the northeast edge of the CKP (Quilty, this volume). The dominant benthic species remains constant through the Upper Cretaceous section at Site 1138.
Planktonic foraminiferal fauna of the Kerguelen Plateau recovered from Sites 1135, 1136, and 1138 have permitted improvement of the biostratigraphic framework for the Southern Ocean region, especially for Turonian-Santonian time (Petrizzo, 2001). Several low- and mid-high-latitude bioevents are useful for correlation across latitudes. Moreover, planktonic foraminiferal assemblages from the Turonian-Coniacian succession throughout the Cretaceous Kerguelen sedimentary sequence clearly indicate affinities with warm-temperate Cretaceous faunas (e.g., Exmouth Plateau), whereas the planktonic fauna in the remainder of the Late Cretaceous section has a typical austral affinity (e.g., Maud Rise and other circum-Antarctic sites) (Petrizzo, 2001).
Sites 1135, 1136, and 1138 on the SKP and CKP (Figs. F1, F2) provide a relatively complete Paleocene and Eocene section of nannofossil-foraminifer oozes and chalks. An apparently complete Cretaceous/Tertiary boundary, in terms of assemblage succession, isotopic signature, and reworking of older (Cretaceous) nannofossil taxa, was recovered at Site 1138 (Arney and Wise, this volume). The boundary is marked by a significant color change, a negative carbon isotope shift, and a nannofossil turnover. As defined above, however, the boundary does not agree with available paleomagnetic data (Shipboard Scientific Party, 2000).
The Paleocene nannofossil assemblage is generally characteristic of high latitudes, and the Southern Ocean biogeography of Hornibrookina indicates a water mass boundary of some kind over the Kerguelen Plateau in the earliest Paleocene (Arney and Wise, this volume). This boundary disappeared by late Paleocene time, which is characterized by warm-water discoasters, sphenoliths, and fasciculiths. This documents relatively equable water temperatures during much of late Paleocene time, and preliminary floral and stable isotope analyses indicate that a reasonably complete record of the Late Paleocene Thermal Maximum (LPTM) was recovered at Site 1135 (Arney and Wise, this volume). At the beginning of middle Eocene time, water temperatures began to decline, nannofossil assemblages became dominated by cool-water species, and discoaster and sphenolith abundances and diversity were curtailed.
High-latitude radiolarians from Site 1138 (Figs. F1, F2) reveal the Oligocene paleoceanographic history of the CKP (Apel et al., this volume). Radiolarians are not preserved in Eocene sediments, but radiolarian preservation improves stepwise through Oligocene toward Miocene time, presumably linked to increased productivity triggered by climatic cooling (Apel et al., this volume). Similar patterns of improving preservation across the Eocene/Oligocene boundary are observed in strata from other DSDP and ODP sites in the Southern Ocean. In contrast to the SKP, however, proxies for productivity are more complex at Site 1138. At the latter, carbonate dissolution is pronounced in the late Eocene-earliest Oligocene section, as indicated by poor preservation of foraminifers and common hiatuses. Radiolarian and diatom preservation, however, does not improve significantly until late early Oligocene time. Multiple measures of radiolarian diversity in Oligocene sediments from Site 1138 closely parallel radiolarian preservation, indicating that productivity controlled preserved radiolarian diversity (Apel et al., this volume). Linear sedimentation rates, calculated from seven diatom bioevents spanning the early Oligocene to middle Miocene section, vary from 9.5 to 18 m/m.y., with an average of 12.6 m/m.y. (Arney et al., this volume).
Site 1139, drilled on the flank of an eroded alkalic volcano, Skiff Bank (Figs. F1, F2), recovered mixed terrigenous-pelagic sediments of Oligocene and early Miocene age. Carbonate profiles through the interval show two prominent minima at ~28 and ~22 Ma that correspond to peaks in terrigenous flux (Reusch, this volume). These and higher-frequency variations of carbonate probably reflect glacioeustatic/climatic changes, with glacial lowstands characterized by lower carbonate percentages, larger grain sizes, and higher opal concentrations. High Oligocene sedimentation rates of ~29 m/m.y. are attributed to high regional pelagic productivity plus the influx of fine terrigenous clastics derived from weathering of the exposed portions of Skiff Bank (Persico et al., this volume). A hostile terrestrial environment, lack of mineral surface area (silt vs. clay), and deposition below the oxygen minimum zone probably account for low organic carbon concentrations through the interval (Reusch, this volume). A paucity of clay suggests that physical as opposed to chemical weathering predominated, and it is likely that Skiff Bank experienced wave and ice erosion, especially during glacial times.
The ~300-m-thick Neogene section recovered at Site 1138 consists primarily of mixed carbonate and biosiliceous clay and ooze, with several thin (1-3 cm) tephra horizons and dispersed tephra (Coffin, Frey, Wallace, et al., 2000). Diatom biostratigraphic datums, supported by nannofossil and planktonic foraminifera biostratigraphy and 40Ar/39Ar ages from ash and tephra horizons, permit construction of a Neogene age-depth model (Bohaty et al., this volume). A possible hiatus exists at the Oligocene/Miocene boundary, and a ~1 m.y. hiatus characterizes the Miocene/Pliocene boundary (Bohaty et al., this volume; see also Vigour and Lazarus, this volume). The tephras at Site 1138 are glass rich, well sorted, and dominantly trachytic to rhyolitic in composition (Bohaty et al., this volume). Volcaniclastic horizons are thought to have originated from Heard Island, ~180 km northwest of Site 1138, and were likely deposited both as primary ash fall and turbidites (Bohaty et al., this volume).
Radiolarian assemblages of late middle Miocene to early Pliocene age are also well preserved at Site 1138 (Figs. F1, F2). Typical Antarctic faunas and consistently good preservation throughout the interval suggest that the site was located within the Antarctic radiolarian province and thus south of the Antarctic Polar Frontal Zone during the interval, which agrees with results from Site 747 to the south (Vigour and Lazarus, this volume). The distribution of other coarse-fraction components (e.g., sponge spicules and lithic fragments), suggests that dissolution, winnowing, and other benthic processes have not significantly affected the assemblages. Despite only moderate drilling recovery of the section, most lower to upper middle Pliocene radiolarian zones are present, although subzones could not be identified. Vigour and Lazarus (this volume) observed a significant discontinuity from 6.1 to 4.6 Ma but could not determine whether it was due to incomplete recovery of the section, to an interval of condensed sedimentation, or to a hiatus. Silicoflagellates are abundant and diverse in the Pliocene and Pleistocene sections, but some intervals of Miocene age at Site 1138 are barren or contain only limited numbers (McCartney et al., this volume). As noted above, linear sedimentation rates calculated from diatom bioevents for the early Oligocene to middle Miocene interval average 12.6 m/m.y. (Arney et al., this volume). At Site 1139 on Skiff Bank, significantly higher Miocene minimum sedimentation rates of ~18 m/m.y. may be explained by high regional pelagic productivity and fine terrigenous clastic input derived from exposed portions of the bank (Persico et al., this volume).