Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Cooper, A. K., P. J. Barrett, H. Stagg, B. Storey, E. Stump, W. Wise, and the 10th ISAES editorial team, eds. (2008). Antarctica: A Keystone in a Changing World. Proceedings of the 10th International Symposium on Antarctic Earth Sciences. Washington, DC: The National Academies Press. A View of Antarctic Ice-Sheet Evolution from Sea-Level and Deep-Sea Isotope Changes During the Late Cretaceous-Cenozoic K. G. Miller,1 J. D. Wright,1 M. E. Katz,1,2 J. V. Browning,1 B. S. Cramer,3 B. S. Wade,4 S. F. Mizintseva1 ABSTRACT INTRODUCTION The imperfect direct record of Antarctic glaciation has led Glacial sediments on the Antarctic continent and its margins to the delayed recognition of the initiation of a continent- (Figure 1) (Barrett, 2007; Barrett et al., 1987; Birkenmajer et sized ice sheet. Early studies interpreted initiation in the al., 2005; Cooper et al., forthcoming; Cooper and OâBrien, middle Miocene (ca 15 Ma). Most current studies place the 2004; Ivany et al., 2006; Kennett and Barker, 1990; Leckie ï¬rst ice sheet in the earliest Oligocene (33.55 Ma), but there and Webb, 1986; LeMasurier and Rex, 1982; Strand et al., is physical evidence for glaciation in the Eocene. Though 2003; Troedson and Riding, 2002; Troedson and Smellie, there are inherent limitations in sea-level and deep-sea iso- 2002; Zachos et al., 1992) provide a direct record of ice tope records, both place constraints on the size and extent sheets, but these records are temporally incomplete and of Late Cretaceous to Cenozoic Antarctic ice sheets. Sea- often poorly dated and thus may not provide a complete and level records argue that small- to medium-size (typically unequivocal history, especially of initiation of ice sheets. 10-12 106 km3) ephemeral ice sheets occurred during the Deep-sea 18O records (Figures 1 and 2) provide well-dated greenhouse world of the Late Cretaceous to middle Eocene. evidence for changes in temperature and 18Oseawater due to Deep-sea 18O records show increases associated with ice-sheet growth, but separating these two effects is difï¬cult many of these greenhouse sea-level falls, consistent with (e.g., Miller et al., 1991). Global sea-level records provide their attribution to ice-sheet growth. Global cooling began evidence for large (tens of meters), rapid (<1 myr) changes in in the middle Eocene and culminated with the major earli- sea level (Figure 2) that can be explained only by changes in est Oligocene (33.55 Ma) growth of a large (25 106 km3) continental ice sheets, though the amplitudes of the changes Antarctic ice sheet that caused a 55-70 m eustatic fall and have been poorly constrained until recently (Miller et al., a 1â° 18O increase. This large ice sheet became a driver of 2005a). Each of these methods has its limitations, but by climate change, not just a response to it, causing increased integrating results from all three we can begin to decipher latitudinal thermal gradients and a spinning up of the oceans the history of Antarctic ice sheets. that, in turn, caused a dramatic reorganization of ocean cir- Over the past 30 years, study of glacial sediments and culation and chemistry. stable isotopes has progressively extended the initiation of ice sheets further back in time. For example, consider the history of Northern Hemisphere ice sheets (NHISs), better 1 Department of Earth and Planetary Sciences, Rutgers University, Pis- known as the âIce Ages.â Glacial deposits formed during cataway, NJ 08854, USA. advances of Laurentide ice led to the mistaken concept of 2 Earth & Environmental Sciences, Rensselaer Polytechnic Institute, Troy, only four Pleistocene glaciations (Flint, 1971), one that was NY 12180, USA. 3 Department of Geological Sciences, University of Oregon, Eugene, OR eventually contradicted by deep-sea 18O records showing 97403, USA. that there were: 4 Now at Department of Geology & Geophysics, Texas A&M University, College Station, TX 77843, USA. 55
56 ANTARCTICA: A KEYSTONE IN A CHANGING WORLD King George 18O Island ?to 20 Ma 2 1.5 1 0.5 0 -0.5 -1 21 6A 22 6AA 6B 23 Zone 24 6C Mi1 25 7 7A 26 8 27 Oi2a? 9 28 10 29 11 Oi2 30 31 12 32 33 Oi1 13 34 35 15 16 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 47 21 48 49 FIGURE 1 Oligocene benthic foraminiferal synthesis compared with the record of glaciomarine sediments (modiï¬ed from Miller et al., 1991). Benthic foraminiferal stable isotope data were stacked and smoothed with a Gaussian convolution ï¬lter in order to remove periods less than 1.0 myr. Because ï¬ltering dampens the amplitude, an arbitrary line was placed through 1.6â° and values higher than this were shaded. 1. Troedson and Smellie (2002); 2. Birkenmajer et al. (2005); 3. Troedson and Riding (2002); 4. Ivany et al. (2006); 5. Strand et al. (2003); 6. Cooper and OâBrien (2004); 7. Kennett and Barker (1990); 8. Zachos et al. (1992); 9. Barrett et al. (1987); 10. LeMasurier and Rex (1982); 11. Leckie and Webb (1986); 12. Barrett (2007).
MILLER ET AL. 57 Chron (C) Sea Level (m) Polarity Epoch Age -150 -100 -50 0 50 100 150 0 1 Pleistocene 2A late Pliocene 3 early 3A 4 late 10 5 middle Miocene 5B =63 m sea-level fall 20 6 early 6C 8 late Oligocene 30 12 early 13 late 16 17 40 18 =50 m sea-level fall 19 middle 20 Eocene 21 50 22 23 early 24 25 late 26 60 Paleocene 27 28 early =40 m sea-level fall 29 ? 30 Maastrichtian 31 70 32 Campanian Error: 33 Â±1 my; Â±15 m ? 80 Late Cretaceous Santonian =25 m sea level fall Coniacian Ice thickness (m) 4000 ? 90 34 Turonian 3500 3000 Cenomanian 2500 2000 100 1500 1000 Early Cretaceous Albian 500 0 =15 m sea level fall 5.0 4.0 3.0 2.0 1.0 0.0 -1.0 -2.0 18 O FIGURE 2 Global sea level (light blue) for the interval 7-100 Ma derived by new backstripped estimates; lowstands are indicated with thin black lines and are unconstrained by data (Miller et al., 2005a). Global sea level (purple) for the interval 0-7 Ma derived from 18O after Miller et al. (2005a). Shown for comparison is a benthic foraminiferal 18O synthesis from 0-100 Ma (red curve) with scale on bottom axis in â° (reported to Cibicidoides values [0.64â° lower than equilibrium]). The portion of the 18O curve from 0-65 Ma is derived using data from Miller et al. (1987); the Late Cretaceous synthesis is after Miller et al. (2004). Data from 7-100 Ma were interpolated to constant 0.1 myr interval and smoothed with a 21-point Gaussian convolution ï¬lter using Igor Proâ¢. Timescale of Berggren et al. (1995). Maps show- ing maximum sizes of ice sheets during peak glaciation for several intervals. Maps are after DeConto and Pollard (2003a,b) and show the amount of equivalent sea-level change proscribed for a given state (e.g., 25 m) and are calibrated to the 18O synthesis (correlation lines) using sea-level data.
58 ANTARCTICA: A KEYSTONE IN A CHANGING WORLD 1. Eight large (~120 m sea-level lowerings), 100-kyr- in Figures 1 and 2) suggest at least three major periods of scale ice ages over the past 800 kyr; Oligocene glaciation. 2. Sixty-two stages5 representing 31 ice-sheet advances A campaign of drilling near Antarctica by the Ocean on ~20-, 41-, and ~100-kyr-scale during the Pleistocene; Drilling Program (ODP) in the late 1980s returned ï¬rm and evidence that supported the 18O record for large ice sheets 3. Over 100 named stages and 50 glacial advances in the earliest Oligocene that included grounded tills and since the late Pliocene âinitiationâ of NHIS (Emiliani, 1955; IRD at lower latitudes than today (see summaries by Miller Hays et al., 1976; Shackleton, 1967). et al., 1991; Zachos et al., 1992). We update the summary of the direct evidence for Eocene-Oligocene ice in the A pulse of ice-rafted detritus (IRD) into the northern North form of tills and glaciomarine sediments near the Antarctic Atlantic ca. 2.6 Ma (late Pliocene) is associated with a major (Figure 1), using more recent drilling by ODP (Cooper 18 O increase; this has been interpreted as the inception of and OâBrien, 2004; Strand et al., 2003), the Cape Roberts NHISs (Shackleton et al., 1984). However, this inception drilling project (Barrett, 2007), plus studies that extend reï¬ects not initiation but an increase in the size of NHISs West Antarctic glaciation back through the early Oligocene growth and decay (e.g., Larsen et al., 1994). Signiï¬cant (Seymour Island) (Ivany et al., 2006) and into the Eocene (at least Greenland-size) NHISs extend back at least to the (King George Island; Birkenmajer et al., 2005; Troedson middle Miocene (ca. 14 Ma; see summary in Wright and and Riding, 2002; Troedson and Smellie, 2002). The evi- Miller, 1996) and recent data indicate that large NHISs may dence for large, grounded ice sheets begins in the earliest have existed since the middle Eocene (Eldrett et al., 2007; Oligocene and continues through the Oligocene (Figure 1). Moran et al., 2006). Seismic stratigraphic studies summarized by Cooper et al. The imperfect direct record of Antarctic glaciation has (forthcoming) also show intense glacial activity beginning similarly led to the progressive extension of initiation of a in the Oligocene in both East and West Antarctica. There is continent-size ice sheet from 15 Ma (middle Miocene) back excellent agreement among proxies that Antarctica was in to 33.55 Ma (earliest Oligocene) (see summaries in Miller et fact an icehouse during the Oligocene and younger inter- al., 1991, 2005a,b; Zachos et al., 1996). In this contribution val. Ice-volume changes have been ï¬rmly linked to global we suggest that continental ice sheets have been intermit- sea-level changes in the Oligocene and younger âicehouse tently present on Antarctica through the Late Cretaceous, a worldâ of large, varying ice sheets (Miller et al., 1998); Pekar time when Antarctica took up residence at the pole (http:// et al. (1996, 2002) recognized that the three to four major www.ig.utexas.edu/research/projects/plates/). Oligocene glaciations of Miller et al. (1991) in fact reï¬ected Deep-sea isotope records have long been used to inter- six myr-scale sea-level falls and attendant ice-growth events. pret Antarctic ice-sheet history. Based on deep-sea 18O The record of glaciomarine sediments documents that the ice records, early studies of Shackleton and Kennett (1975) sheets occurred in Antarctica (Figure 1), though an NHIS and Savin et al. (1975) assumed that a continent-size ice component cannot be precluded due to scarce Northern sheet ï¬rst appeared in Antarctica in the middle Miocene Hemisphere Oligocene records. Today 33.5 Ma is cited as the (ca. 15 Ma), though they noted that glaciation (in the form inception of the Antarctic ice sheet, though this supposition of mountain glaciers and sea ice) probably occurred back is now being challenged and pushed back into the Cretaceous through the Oligocene. Also using isotope data, Matthews (Miller et al., 1999, 2003, 2005a,b; Stoll and Schrag, 1996). and Poore (1980) suggested that large ice sheets existed in Nevertheless, 33.55 Ma was probably the ï¬rst time in the Antarctica since at least the earliest Oligocene (33.5 Ma). past 100 myr that the ice sheets reached the coast, allowing The differences in interpretation partly illustrate problems large icebergs to calve and reach distal locations such as in using 18O as an ice-volume proxy, because deep-sea the Kerguelen Plateau (Figure 1) (ODP Site 748) (Zachos 18 O values also reï¬ect deep-water temperature changes et al., 1992). that generally mimic high-latitude surface temperatures. There is evidence for glaciation in the older Antarctic Miller and Fairbanks (1983, 1985) and Miller et al. (1987, record. Coring by ODP Legs 119 and 120 (Barron and Larsen, 1991) provided the strongest isotopic evidence for the pres- 1989; Breza and Wise, 1992) and seismic stratigraphic stud- ence of ice sheets during the Oligocene; high 18O values ies (Cooper et al., forthcoming) suggest the possibility of late measured in deep-sea cores (>1.8â° in Cibicidoides spp. or Eocene (or even possibly middle Eocene) glaciers in Prydz >2.4â° in Uvigerina spp.) require bottom-water temperatures Bay. Seismic stratigraphic studies (Cooper et al., forthcom- colder than today if an ice-free world is assumed. Such low ing) also suggest the possibility of late Eocene glaciers in the bottom-water temperatures are incompatible with an ice- Ross Sea. Other studies extend the record for West Antarctic free world; their isotopic synthesis (updated and presented glaciation back from 10 Ma to 45 Ma (Birkenmajer, 1991; Birkenmajer et al., 2005). Though Birkenmajer et al. (2005) 5 Although called âoxygen isotopic stagesâ by paleoceanographers for interpreted the Eocene tills as evidence for mountain glaciers decades, the term âstageâ is a stratigraphic term reserved for characterizing and not necessarily ice sheets, it points to the likelihood that time-rock units (Hedberg, 1976). The proper term for âisotopic variationsâ are âzones in depth and chrons in time.â the continental interior could have supported an ice sheet
MILLER ET AL. 59 in the middle Eocene. Margolis and Kennett (1971), Wei Miller et al., 2005a) and published isotope data (Figures 2 (1992), and Wise et al. (1991) interpreted middle Eocene and 3), we argue for the likely presence of small- to medium- quartz grains in Eltanin cores from near Antarctica as reï¬ect- size ephemeral ice sheets in the greenhouse world of the ing IRD (Figure 1), though the evidence of this as IRD versus Late Cretaceous to Eocene. Though there were ephemeral other transport mechanisms is not compelling. ice sheets in the greenhouse world, the Eocene-Oligocene Early studies recognized the importance of a cold, if transition represented the beginning of the icehouse with the not fully glaciated, Antarctica on deep-water circulation largest cooling event of the last 100 myr, one that resulted (Kennett, 1977; Kennett and Shackleton, 1976; Shackleton in an ice sheet that reached the coast for the ï¬rst time. We and Kennett, 1975). Kennett (1977) attributed the Eocene- review published and recently submitted evidence for the Oligocene transition to the development of a nascent Ant- nature and timing of paleoceanographic changes associated arctic Circumpolar Current that caused thermal isolation of with the Eocene-Oligocene transition (Figure 4) and pres- Antarctica, development of sea ice (though not continental- ent new and published comparisons of global deep-water scale glaciation), and an increase in Antarctic bottom water changes that resulted from this glaciation and attendant (AABW). The formation of AABW today is particularly cooling (Figures 4 and 5). sensitive to sea ice, and geologic evidence is clear that an erosional pulse of deep water in the Southern Ocean occurred THE CASE FOR ICE SHEETS IN THE GREENHOUSE near the Eocene-Oligocene transition (Kennett, 1977; Wright WORLD (CRETACEOUS-EOCENE) and Miller, 1996). It was also known that a major drop in the calcite compensation depth (CCD) occurred across the The sea-level curves of Exxon Production Research Com- Eocene-Oligocene transition (van Andel, 1975) in concert pany (EPR) (Haq et al., 1987; Vail et al., 1977) stimulated with this change in deep-water circulation. However, our interest in large, rapid, global sea-level changes. Vail et al. understanding of deep-water history and its relationship to (1977) reported numerous large (>100 m) Phanerozoic sea- the evolution of Antarctic ice volume has been unclear in level falls, including a 400 m drop in mid-Oligocene. On the part because of uncertainties in the proxies for deep water Vail curve these falls were shown as virtually instantaneous. and continental ice sheets. Subsequent study showed that this saw-toothed pattern with Here we review interpretations of Antarctic ice-volume extremely rapid falls was an artifact of measurement of changes using sea-level and oxygen isotopic records. Using coastal onlap versus ofï¬ap (Thorne and Watts, 1984), but recently published sea-level curves (Kominz et al., in review; subsequent generations of the EPR curve covering the past A B FIGURE 3 (A) Planktonic (red line) and benthic (blue line) 18O record of Moriya et al. (2007) plotted with a full 4.5â° range. Benthic foraminifera 18O is based on the combined records of Bolivina anambra, Gavelinella spp., and Neobulimina spp. species. (B) Moriya et al.âs (2007) 18O record of planktonic (red line) and benthic foraminifera Bolivina anambra (blue line) plotted on enlarged scales with 1.5â° rang- es; note different scales for benthic (bottom scale) and planktonic (top scale) values. Only values of Bolivina anambra greater than â2â° are included. Also shown is the sea-level record (black line) of Kominz et al. (in review) shifted in age by ~0.2 myr to maximize correlations.
60 ANTARCTICA: A KEYSTONE IN A CHANGING WORLD FIGURE 4 Comparison of continental margin records from New Jersey and Alabama (Miller et al., 2008), global sea-level estimates, and benthic foraminiferal 18O records from deep Paciï¬c ODP Site 1218 (Coxall et al., 2005) and St. Stephens Quarry, Alabama (Miller et al., 2004). Shown at right is a blowup of ODP Site 1218 oxygen isotopes that includes carbonate percentage and mass accumulation rate (AR) 2004 data (Coxallright is2005) that of ODP Site 1218 oxygen isotopes that includes carbonate percentage and mass accumulation reï¬ected in the ). Shown at et al., a blowup shows two large drops associated with the precursor and Oi1 oxygen isotope increases that are rate (AR) data core photograph at right (light colorlarge drops associated with the precursor and Oi1 oxygen isotope increases that are reï¬ected in the core (Coxall et al., 2005) that shows two = carbonate rich; dark = carbonate poor). photograph at right (light color = carbonate rich; dark = carbonate poor). 200 myr still showed 100 m falls in much less than 1 myr but the effect is small (i.e., a 10Â°C global warming would (Haq et al., 1987), including a mid-Oligocene fall of ~160 m. cause only a 10 m sea-level rise [Pitman and Golovchenko, Though the EPR curves have been strongly criticized for 1983]). Changes in terrestrial storage in lakes and ground- their methodology and proprietary data (Christie-Blick et al., water can only explain 5 m of sea-level change (Pitman and 1990; Miall, 1991), recently published sea-level estimates Golovchenko, 1983). Desiccating and reï¬lling Mediterra- (Kominz et al., in review; Miller et al., 1998, 2005a) show nean basins could explain very rapid (<1 kyr) changes, but that the timing of the EPR curves is largely correct. These this effect is small (~10 m) and it is impossible to explain the recent estimates (discussed below) show that the EPR sea- number of Late Cretaceous to early Eocene sea-level events level amplitudes are typically two to three times too high with this mechanism (Pitman and Golovchenko, 1983). Thus but still require tens of meters of change in much less than only ice-volume changes can explain these large (tens of 1 myr. meters) sea-level changes, even in the supposedly ice-free Such rapid sea-level changes pose an enigma, because world of the Late Cretaceous to Eocene. the only known mechanism for causing sea-level changes in Matthews and Poore (1980) ï¬rst realized this enigma excess of 10 m in less than 1 myr is glacioeustasy (Pitman and Matthews (1984), based on sea-level records and his and Golovchenko, 1983). No other known mechanisms reinterpretation of the 18O record, postulated that intermit- (steric effects, storage in lakes, deep-water changes, ground- tent ice sheets occurred in the mid-Cretaceous through the water, or sea ice) can explain these changes (see Figure 1 Paleogene. Based on a comparison of continental margin in Miller et al., 2005a). Temperature changes can explain and 18O records, Miller et al. (1987, 1991) suggested that rapid sea-level changes such as the changes happening today, the growth and decay of continental ice sheets controlled
MILLER ET AL. 61 FIGURE 5 Comparison of benthic foraminiferal 18O data from Paciï¬c ODP Site 1218 (Coxall et al., 2005) with the percentage of northern component water (NCW) (Wright and Miller, 1996) and unpublished difference curve between a Paciï¬c and southern ocean 18O values. Values were interpolated to constant 0.1 myr intervals and smoothed with an 11-point Gaussian convolution ï¬lter using Igor Pro Â®. Data were divided into Maud Rise sites closest to Antarctica and subantarctic sites. Timescale after Berggren et al. (1995). sea-level changes in the Oligocene. Subsequent studies have Stoll and Schrag (2000) examined mid-Cretaceous isotopic conï¬rmed the existence of Oligocene ice sheets (see Figure variations and revived Matthewsâs idea by postulating that 1, and summaries of Miller et al., 1991; Zachos et al., 1994) continental ice sheets grew and decayed in this greenhouse and led to the suggestion that small- to medium-size ice world, and recent sea-level studies have supported their sup- sheets existed in the middle to late Eocene (Browning et al., position (Kominz et al., in review; Miller et al., 2005a; Van 1996). Yet other than Matthews (1984) the concept of ice Sickel et al., 2004). sheets during the interval of peak global warmth, the green- Studies from the New Jersey coastal plain provided a house world of the Late Cretaceous to Eocene, was ignored. eustatic estimate for the past 100 myr, using inverse models
62 ANTARCTICA: A KEYSTONE IN A CHANGING WORLD termed âbackstrippingâ (Kominz et al., 1998, in review; (2003, 2004, 2005a,b) postulated the existence of ice sheets Miller et al., 2005a; Van Sickel et al., 2004). Backstripping in the greenhouse world of the Late Cretaceous to Eocene. progressively removes the effects of sediment compaction Miller et al. (2005b) presented a new view of Earthâs cryo- and loading from observed basin subsidence (e.g., Kominz spheric evolution that reconciled warm, largely ice-free poles et al., 1998) and thermal-ï¬exural subsidence is modeled by with cold periods that resulted in glacioeustatic lowerings. ï¬tting exponential curves to the remaining observed subsid- Their view was developed from work by DeConto and Pol- ence. The difference between the best ï¬t exponential curve lard (2003a,b) who used a coupled global climate and ice- and subsidence is a result of either eustatic change or any sheet model that accounts for bedrock loading, lapse rate subsidence unrelated to thermal subsidence (Kominz et al., effects of topography, surface mass balance, basal melting, 1998). We have applied this technique to coreholes from the and ice ï¬ow and computes both the sea level and 18O effects New Jersey and Delaware coastal plains by dating sequences of ice growth. The DeConto and Pollard (2003a,b) model (unconformity bounded units) with a Â±0.5 myr or better reso- used declining atmospheric CO2 values of ~3 to 2 times lution and developing a water depth history by integrating present day, similar to empirical estimates for declining CO2 biofacies and lithofacies analysis. The similarity of records from the Eocene to Oligocene (Pagani et al., 2005). These (six sites for the Cenozoic and four for the Cretaceous) model runs were used to estimate the size and geographic (Miller et al., 2004) indicates that the effects of thermal distribution of ice sheets (Figure 2) for Oligocene continental subsidence, loading, and water-depth variations have been conï¬gurations, though the modeling results are applicable to successfully removed. Backstripping, seismicity, seismic the older record as long as Antarctica was a polar continent stratigraphic data, and distribution patterns of sediments all (i.e., for the entire interval considered here). indicate minimal tectonic effects on the Late Cretaceous to We estimate Antarctic ice volume using the New Jersey Tertiary New Jersey coastal plain (Miller et al., 2004, 2005a), sea-level record (Figure 2). The typical 15-30 m green- though some minor (10-30 m) differences between New house eustatic falls correspond to growth of ice volumes of Jersey and Delaware must be attributed to local subsidence 8-12 106 km3 using a 0.1â°/10 m calibration of sea level or uplift (Browning et al., 2006). and ice volume (DeConto and Pollard, 2003a,b). The excep- The timing of sequences in New Jersey and other mar- tion is the large Campanian/Maastrichtian boundary fall of gins is similar, suggesting a global cause. There are few other 40 m that suggests growth of an ice sheet of 17 106 km3. backstripped records to test the New Jersey eustatic estimate As illustrated in Figure 2, these ice sheets did not reach the against, though a Russian platform backstripped eustatic Antarctic coast. This together with the fact that this ice sheet estimate (Sahagian et al., 1996) shows remarkably similar probably existed for relatively short periods (see below) changes to New Jersey in the interval of overlap from 100 reconciles ice with coastal warmth in Antarctica. Though the Ma to 90 Ma (Miller et al., 2003, 2005a; Van Sickel et al., temporal coverage is limited, there is ample evidence that 2004). In addition, comparison of the New Jersey record with Antarctic coastal climates were quite warm during much of northwest Europe (Ali and Hailwood, 1995; Hancock, 1993), the Late Cretaceous to Eocene (e.g., Askin, 1989; Francis and the U.S. Gulf Coast (Mancini and Tew, 1995), the EPR syn- Poole, 2002); deep waters were relatively warm also (e.g., thesis (Haq et al., 1987), and long-term sea-level predictions early Maastrichtian paleotemperatures of ~10Â°C at 1500 m from Milankovitch forcing (Matthews and Frohlich, 2002) paleodepth on the Maud Rise) (Barrera and Huber, 1990). suggests that Late Cretaceous to Eocene sea-level falls were Yet these coastal and offshore studies do not address regions rapid, synchronous, and global (see summary ï¬gures in of ice sheet nucleation predicted for the continental interiors Miller et al., 2004). Therefore, it is unlikely that the New (Figure 2) (DeConto and Pollard, 2003a,b). Jersey eustatic estimate can be attributed to tectonics. The These ice sheets were ephemeral and for the most part only mechanism than can explain the observed rates of sea- Antarctica lacked ice sheets during the greenhouse world level change (in excess of 25 m/myr) is the growth and decay except for these cool or cold periods, called âcool snapsâ of continental ice sheets that caused glacioeustatic changes. by Royer et al. (2004). It is not possible to say how long Error bars for the sea-level falls are discussed in detail by these cool or cold intervals lasted. Matthews and Frohlich Kominz et al. (in review). The greatest uncertainty is that the (2002) computed an estimated Cretaceous sea-level and ice- onshore sites mostly miss the lowstands and thus are minimal volume curve based on Milankovitch forcing, and minima estimates of the eustatic falls. The Late Cretaceous to Eocene in this curve appear to correlate with sequence boundaries falls were typically 25 m, with small events (15 m) near the in New Jersey (Miller et al., 2005b). The dominant beats detection level. The only exception was a large Campanian/ in the predicted curve are the ca. 2.4 myr and 405 kyr very Maastrichtian boundary (ca. 71.5 Ma) sea-level fall of 40 m. long and long eccentricity cycles and the 1.2 myr tilt cycle. It should be noted that the New Jersey eustatic estimate must Yet these longer-term modulations predict that shorter-term be considered a testable model. A true global sea-level curve periods (tilt and precession) must have been operative and it must be derived from numerous margins, not just one; thus seems likely that the cool intervals only lasted during minima further studies are needed to validate this record. in insolation on the precession and tilt periods. Studies of Based on the New Jersey sea-level records, Miller et al. myr-scale sea-level events in the Oligocene suggest that peak
MILLER ET AL. 63 cold intervals lasted only 2-3 tilt (41 kyr) cycles (Coxall et late Eocene ice sheets. Cretaceous to early Eocene 18O data al., 2005; Zachos et al., 1996). Thus we speculate that ice are sparse because of poor core recovery and diagenesis (i.e., sheets existed only for brief intervals (~100 kyr for a typical isotopic studies of sections with >400 m burial are suspect) 2.4 myr cycle) during the greenhouse. (1987), but some comparisons can be made. The large Cam- We illustrated Antarctic cryospheric evolution using panian/Maastrichtian boundary sea-level fall is associated maps derived from the models of DeConto and Pollard with a major 18O increase in both planktonic and benthic (2003b). These maps show areas and thicknesses computed foraminifera (Miller et al., 1999, using data of Barrera and from the models, which also compute the equivalent water Savin, 1999; and Huber et al., 2002). Also, large (>0.75â°) volume and sea level. We correlate each map and its atten- mid-Cenomanian (ca. 96 Ma) and mid-Turonian (ca. 92-93 dant sea-level fall to equivalent sea-level falls estimated in Ma) benthic foraminiferal 18O increases reported from west- New Jersey (e.g., the map for a 40 m fall is aligned with ern North Atlantic ODP Site 1050 (Huber et al., 2002) cor- the 40 m 71.5 Ma fall) (Figure 2). For periods of small (~15 relate with major sea-level falls in New Jersey. These large 18 m) sea-level falls that typiï¬ed most of the Late Cretaceous O increases at 92-93 Ma and 96 Ma cannot be entirely and early Eocene, models indicate that small isolated ice attributed to ice-volume changes because this would require caps would have formed in the highest elevation of Dron- ice sheets larger than those of modern times, and in fact ning Maud Land, the Gamburtsev Plateau, and the Trans- much of the 18O signal must be attributed to deep-sea (hence antarctic Mountains (Figure 2) under high CO2 conditions. inferred high-latitude; see below) temperature change. Miller For periods with moderate sea-level falls (25 m) that were et al. (2005b) estimated that about two-thirds of the 18O typical of the larger events in greenhouse intervals (e.g., signal was attributable to temperature, suggesting about the mid-Turonian and mid-Cenomanian), a small ice-sheet 0.25â° was due to a change in 18Oseawater, corresponding to a threshold was reached owing to height and mass balance sea-level change of ~25 m and growth of about one-third of feedback; the three ice sheets (Dronning Maud Land, the the Antarctic ice sheet. Gamburtsev Plateau, and the Transantarctic Mountain) con- A recent study has challenged correlations of sea-level tinued to grow but not coalesce, and were isolated from the falls and 18O increases for the Cenomanian. Moriya et coast. With a 40 m sea-level lowering, ice caps would have al. (2007) tested evidence for Cenomanian glacioeustasy begun to coalesce (Figure 2); this conï¬guration would have by generating high-resolution (26 kyr sampling) benthic been achieved only during the largest of the greenhouse and planktonic 18O records from ODP Site 1258 on the sea-level falls (e.g., the 71.5 Ma Campanian/Maastrichtian Demerara Rise (western tropical Atlantic Ocean). They con- event). With sea-level falls of over 50 m, the three ice-sheet cluded that there was no support for Cretaceous glaciation nodes would have united (Figure 2); this conï¬guration and called into question evidence for greenhouse ice sheets. would not have been achieved until CO2 fell below a critical We replotted their data (Figure 3) and make the following threshold (estimated at 2.8 times preanthropogenic levels) observations: (DeConto and Pollard, 2003a,b) in the earliest Oligocene (33.55 Ma). 1. Benthic foraminiferal 18O values from the Dem- The elevation of Antarctica in the greenhouse world is erara Rise (Moriya et al., 2007) show very large amplitudes an important unknown. The models of DeConto and Pollard (up to 3â°) that would require deep-water temperature (2003a) account for changes in elevation due to isostasy changes of 13Â°C in a few 100 kyr (Figure 3). The extremely and mountain building and are appropriate for Oligocene low benthic 18O values (less than â4â°, mean value of and later Eocene conï¬gurations. However, the elevation of â1.8â°) correspond to maximum and average deep-water the continent may have been lower in the earliest Eocene temperatures of 30Â°C and 19.5Â°C, respectively (assuming and older. Uplift of the Transantarctic Mountains may have an ice-free 18Oseawater value of â1.2â°) (Moriya et al., 2007). begun in the Cretaceous (Fitzgerald, 2002), though it is also Considering that the lowest benthic foraminiferal 18O values possible that uplift did not begin until the Eocene (ten Brink for the Cenomanian at western North Atlantic ODP Site 1050 et al., 1997). We concede that it may have been more difï¬cult (Figure 2) (Huber et al., 2002) are â1.4â° to â1.6â°, corre- to nucleate ice sheets on a lower continent but maintain that sponding to deep-water temperatures of 18-19Â°C, we suggest the elevational history of Antarctica is poorly known. that Moriya et al.âs (2007) benthic foraminiferal records The scenario for Late Cretaceous to Cenozoic ice-sheet are overprinted by diagenesis, very large local 18Oseawater history (Figure 2) is partly testable with 18O data because effects, or by analysis of multiple species with very large each sea-level event should be associated with a 18O vital effects. Their 18O values on the genus Neobulimina increase due to ice growth. This prediction has been veriï¬ed particularly show considerable variations, with many values for the Oligocene to middle Miocene (Miller et al., 1996, close to those of planktonic foraminifera. We culled their 1998) and extended back to the middle to late Eocene, where benthic foraminiferal 18O, accepting only Bolivina data and sea-level changes are coupled with 18O increases (Browning rejecting all values lower than â2â° as physically unrealistic; et al., 1996), suggesting glacioeustatic control. Tripati et al. the resulting dataset shows a similar pattern to the planktonic 18 (2005) also used stable isotopic data to argue for middle and O record (Figure 3b).
64 ANTARCTICA: A KEYSTONE IN A CHANGING WORLD 2. Though Moriya et al. (2007) claimed minimal sion below). If fully compensated by isostasy, this 55-70 m change in planktonic foraminiferal 18O values, this is represents ~82-105 m of change in water volume or apparent largely an artifact of the scale of 5â° on which the data were sea level (Pekar et al., 2002), which would correspond with plotted (e.g., Figure 3a). Plotting planktonic foraminiferal 115-150 percent of the modern Antarctic ice sheet or 100-130 18 O values with a 1.5â° scale (Figure 3b) show that values percent of the entire modern ice inventory. increase by ~0.5-0.6â° in the mid-Cenomanian (~95.5 Ma) These large ice estimates require storage of ice outside at precisely the same time as the sea-level fall delineated in East Antarctica. Though earlier studies assumed that the West New Jersey (Figure 3). Antarctic ice sheet did not develop until the later Miocene 3. As noted by Miller et al. (2005a) and Moriya et (Kennett, 1977), recent studies (Ivany et al., 2006) suggest al. (2007), 18Oseawater effects for these greenhouse ice sheets that glacial ice extended to sea level in this region by the ear- would have been small (~0.2-0.3â°), but this is detectable liest Oligocene; these were not just mountain glaciers. This within measurement error. While we have observed and is supported by seismic stratigraphic evidence summarized would reasonably expect cooler deep-water temperatures by Cooper et al. (forthcoming) for glacial inï¬uences in both associated with these greenhouse ice-growth events (e.g., East and West Antarctica beginning in the Oligocene. two-thirds of the Campanian/Maastrichtian fall must be These sea-level changes suggest at least moderate due to cooling), there is no reason to require proportional glaciation in the Northern Hemisphere in the earliest coupling of cooling and ice volume as suggested by Moriya Oligocene. Today the Greenland ice sheet is composed of et al. (2007). In fact, we suspect that one reason that early 6.5 m (Williams and Ferrigno, 1995) of sea-level equiva- Eocene ice-volume events have proven to be so elusive is lent. Tripati et al. (2005) have argued for very large (tens that there was little change in deep-water temperature and of meters) NHISs in the earliest Oligocene. They attribute the ice-volume signal is quite small (0.2-0.3â°). the entire 18O increase observed in the deep Paciï¬c to 4. Thus the Demerera Rise isotope data are consistent growth of ice sheets because Mg/Ca data associated with with (rather than disprove) the idea that there was a seawater the Oi1 18O increase. Mg/Ca data for Oi1 time indicate a 18 O change on the order of ~0.2-0.3â° associated with the 2Â°C warming that is an artifact of carbonate undersatura- mid-Cenomanian event. tion in the deep sea (Lear and Rosenthal, 2006). However, though there is evidence for localized Northern Hemisphere We conclude that sea-level records indicate that small- glaciation in the Paleogene (Eldrett et al., 2007; Moran et to medium-size (10-15 106 km3), ephemeral (lasting a few al., 2006), the ï¬rst appearance of IRD in the North Atlantic tilt cycles based on analogy with high-resolution Oligocene outside the Norwegian-Greenland Sea did not occur until studies of Zachos et al., 1994) ice sheets occurred during the 2.6 Ma (Shackleton et al., 1984). The Eocene IRD evidence greenhouse world of the Late Cretaceous to middle Eocene. presented by Eldrett et al. (2007) and Moran et al. (2006) However, an Antarctic record of these glaciations exists only is compelling but limited to the high-latitude Norwegian- from the middle Eocene (Figure 1). Greenland Sea and Arctic and does not indicate widespread Northern Hemisphere glaciation. We thus doubt the pres- ence of very large (tens of meters equivalent) ice sheets in THE BIG CHILL INTO THE ICEHOUSE: the Northern Hemisphere prior to 2.6 Ma. THE EOCENE-OLIGOCENE TRANSITION The suggestion that the Antarctic ice sheet was as large Following peak warmth in the early Eocene (ca. 50 Ma), if not larger than today needs to be reconciled with paleo- benthic foraminiferal deep-water and by extension high-lati- botanical data that suggest that ice-free refugia are needed tude surface waters cooled across the early to middle Eocene for the Oligocene (e.g., Francis and Poole, 2002). Drilling boundary (ca. 48-49 Ma) and in the late middle Eocene (ca. in the Weddell Sea highlights this enigma; lowermost Oli- 44-41 Ma) (Figure 2). The sea-level record shows develop- gocene grounded diamictons are overlain by strata contain- ment of progressively larger ice sheets in the middle to late ing Nothofagus leaves (Kennett and Barker, 1990). This is Eocene (Figure 2). The Eocene-Oligocene transition culmi- consistent with our scenario based on sea-level and stable nated with the largest 18O increase of the past 50 myr, the isotopic studies (Coxall et al., 2005; Zachos et al., 1996) earliest Oligocene Oi1 18O maximum (33.55 Ma). that earliest Oligocene ice sheets formed rapidly (<< 1 myr) Antarctica entered the icehouse in the earliest Oligocene in a series of Milankovitch tilt-driven (41 kyr) increases that (33.55 Ma). There is widespread evidence for large ice sheets lasted for a few hundred kyr, and then almost completely col- (IRD and grounded diamictons [Figure 1]) and we suggest lapsed again, only to reform again at Oi1a time (ca. 33 Ma), that Antarctica was nearly fully glaciated. A eustatic fall of but it leaves open the question of where these refugia were 55-70 m occurred at 33.55 Ma in association with the 18O located during times of maximum glaciation. Under true increase. This corresponds to an Antarctic ice sheet that was icehouse conditions that began in the earliest Oligocene, the ~80-100 percent of the size of the present-day East Antarctic ice sheet reached the coastline (Figure 2), which limited its ice sheet (today a sea-level equivalent of ~63 m is stored in further growth. The sea-level estimates suggest a minimum East Antarctica and 5-7 m in West Antarctica; see discus- drop of 55 m, which is illustrated by calving along much of
MILLER ET AL. 65 the coastline except Wilkes Land (Figure 2); however, our AABW, with a large one at the Eocene-Oligocene boundary. upper limit eustatic lowering of 70-105 m would require In the North Atlantic a major erosional pulse was associ- ice volumes exceeding modern, and thus reaching the coast ated with reï¬ectors R4 (Miller and Tucholke, 1983) and along much of the margin. Au (Tucholke and Mountain, 1979), both of which date to Using data of Miller et al. (2008), we present a detailed near the Eocene-Oligocene boundary (Mountain and Miller, view of the climate changes that spanned the Eocene- 1992). Northern Hemisphere high latitudes cooled concomi- Oligocene boundary (Figure 4). The Eocene-Oligocene tantly with polar cooling in the Southern Hemisphere, mak- transition is associated with a long-term (10 7 yr scale) CO2 ing the North Atlantic a viable source region for deep water drawdown (Pagani et al., 2005) and related temperature (albeit brieï¬y) in the earliest Oligocene. change that triggered a precursor 18O increase at 33.8 Ma Stable isotope reconstructions are generally consistent of ~0.5â° identiï¬ed at Paciï¬c ODP Site 1218 (Coxall et with the scenario derived from mapping of hiatuses and al., 2005) and shelf site at St. Stephens Quarry, Alabama seismic disconformities. Carbon isotopic reconstructions of (Miller et al., 2008), followed by a 1.0â° 18O increase at the Oligocene show low vertical and interbasinal differences 33.55 Ma (= Oi1, earliest Oligocene). This is consistent (Diester-Haass and Zachos, 2003; Miller and Fairbanks, with models that predict climate response of at least two 1985; Salamy and Zachos, 1999; Wade and PÃ¤like, 2004), nonlinear jumps associated with the transition (DeConto limiting the use of this proxy for reconstructing deep-water and Pollard, 2003b). The precursor increase is not associ- changes. We attribute these low gradients to a general ated with an observable change in sea level (i.e., a change decrease in export production; although the spin-up of the greater than 15-20 m), suggesting that it was caused by oceans increased export productivity in eutrophic areas a 2Â°C cooling, not ice-sheet growth (Miller et al., 2008). (Moore et al., 2004), available nutrients limited general Global sea level dropped by ~55-82 m at 33.55 Ma, indi- export production, which was remarkably low in oligotrophic cating that the deep-sea 18O increase was due to transfer areas (Miller and Katz, 1987). Despite the low Oligocene 13 of the water from the oceans to the Antarctic ice sheet, C sensitivity (and the resulting ï¬at line on Figure 5), 13 which was 80-130 percent of the modern size, as discussed C records indicate that there was a short early Oligocene above. The amount of cooling associated with Oi1 is still interval (ca. 33-32 Ma) with high Atlantic-Paciï¬c 13C gradi- uncertain. Using the 55 m fall and a calibration of 0.1â°/10 ents (Miller, 1992); Wright and Miller (1996) quantiï¬ed the m (DeConto and Pollard, 2003b) suggests that deep-water percentage of northern component water using interbasinal 13 temperatures dropped 2Â°C; using the higher sea-level fall C gradients (Figure 5). (82 m) suggests only 1Â°C of cooling. Oxygen isotopic records indicate that the Antarctic was The Antarctic ice sheet reached the coastline for the the source of water with high 18O values in the late middle first time in the earliest Oligocene (Figure 2) and this to late Eocene (Figure 5). ODP Site 689 on the Maud Rise large ice sheet became a driver of, not just a response to, (1800 m paleodepth) (Kennett and Stott, 1990; Thomas, climate change. The earliest Oligocene was characterized 1990) records very high 18O values beginning with the late by increased latitudinal thermal gradients (Kennett, 1977). middle Eocene cooling and staying high through the Oligo- This increase caused: (1) enhanced wind intensities and a cene (Figure 5). This âcold spigotâ indicates that AABW spinning up of the oceans, resulting in increased upwell- had a strong cold inï¬uence close to the continent, but this ing (Suko, 2006) and increased thermohaline circulation, inï¬uence was reduced by mixing away from the proximal with transient erosional pulses of North Atlantic deep water southern ocean (Miller, 1992). Subantarctic 18O records (Miller and Tucholke, 1983; Wright and Miller, 1996) and show that the cold water mass did not strongly inï¬uence AABW (Kennett, 1977; Wright and Miller, 1996); (2) a regions further from the continent until the end of the Eocene major drop in the CCD as the deep oceans cooled, became (Figure 5). Very high subantarctic 18O values during the better ventilated, and had reduced residence time and acidity Eocene-Oligocene transition suggest a pulse of AABW that (Figure 4) (Coxall et al., 2005; van Andel, 1975); and (3) a peaked with the Oi1 glaciation (Figure 5), consistent with the large increase in diatom diversity due to intensiï¬ed latitudi- erosional record (Kennett, 1977; Wright and Miller, 1996). nal thermal gradients and upwelling (Falkowski et al., 2004; This increase in thermohaline circulation is coincident with Finkel et al., 2005; Pak and Miller, 1992). the drop in the calcite compensation depth. Seismic stratigraphy and mapping of hiatuses indicate a The drop in the CCD was caused by increased thermo- major strengthening of thermohaline circulation in the Oli- haline circulation that caused a decrease in oceanic resi- gocene. The Antarctic has generally been a source of deep dence time and a decrease in deep-ocean acidity. Coxall et water to the oceans (analogous to AABW) throughout much al. (2005) showed that the CCD drop occurred in two steps of the later Cretaceous and the Cenozoic (Pak and Miller, associated with the precursor and 18O increases (Figure 1992; Mountain and Miller, 1992). The Eocene-Oligocene 4); the drop in percentage of carbonate at the precursor boundary saw intensiï¬cation of AABW. Kennett (1977) and event was as large or larger than at Oi1 time (Figure 5), Wright and Miller (1996) showed that numerous hiatuses but accumulation rate data indicate that the latter event was in the southern ocean are attributable to erosional pulses of more signiï¬cant. The cause of the drop in the CCD has been
66 ANTARCTICA: A KEYSTONE IN A CHANGING WORLD debatable. Any deepening of the CCD must be attributed to integrating previous reviews of sea-level change (Miller et one of the following: al., 2005a), greenhouse ice sheets (Miller et al., 2005b), new data from the Eocene-Oligocene transition (Miller et al., 1. An increase in deep-basin carbonate deposition at 2008), and deep-sea circulation changes (Wright and Miller, the expense of shallow carbonates (shelf-basin fractionation, 1996). The sea-level record was recently updated (Figure 2) as invoked by Coxall et al., 2005) for this drop. We argue by Kominz et al. (in review) who presented a thorough error that shelf-basin fractionation cannot be invoked to explain analysis. We can make certain statements about sea-level the CCD drop associated with the precursor 18O increase, changes: because there is little or no sea-level fall during the precur- sor; in addition, they note that the amount of eustatic fall 1. Very high-amplitude, myr-scale sea-level changes (55-82 m) associated with the 33.55 Ma Oi1 18O increase (up to 160 m) of EPR are not supported. is too small to account for the large deepening of the CCD. 2. Our sea-level record (Figure 2) is derived from stud- 2. A global intensiï¬cation of carbonate export to the ies of only one margin, except for the interval from 90 Ma deep sea caused by increased continental input to the oceans. to 100 Ma, which is corroborated by the Russian Platform Weathering rates increased in the earliest Oligocene (Robert (Sahagian et al., 1996). A true global sea-level curve must and Kennett, 1997) and may have contributed to a global be derived from numerous margins and thus the curve pre- increase in carbonate production. However, global Oligocene sented must be considered a testable model. Further studies sedimentation rates were low (Thunell and Corliss, 1986), are needed to validate this record. and it is doubtful that input changes could explain the CCD 3. Our best estimate of greenhouse sea-level ampli- drop. Rea and Lyle (2005) argued that the CCD drop could tudes are 15-25 m on the myr scale, except for the ca. 71.5 not be entirely due to shelf-basin shift and suggested that a Ma event, which was over 40 m. Icehouse amplitudes are as sudden increase in weathering and erosion rates is unlikely to high as 55-70 m. We do not capture most lowstands in our account for the change, thus implicating changes in deep-sea sea-level record, and these estimates are minima. preservation. 4. The New Jersey estimates are eustatic, having been 3. A large deep-water cooling. Cooling may have con- corrected for the effects of loading. To estimate the actual tributed to the CCD falls, but the minor amount of cooling volume of ice equivalent requires the making of assumptions at Oi1 time (1-2Â°C as discussed above) cannot fully explain about isostatic effects due to water loading. As noted by this drop. Pekar et al. (2002) a eustatic change of 55 m would require 4. A decrease of deep-ocean residence time. The a change in water volume of 82 m, assuming full compensa- links among deep-sea temperature, an increase in thermo- tion. Similarly, in the younger record the change from 18 ka haline circulation, and the drop in the CCD (Figure 4) are to present of 120 m measured in Barbados (Fairbanks, 1989) manifested as to the cause: a decrease in ocean acidity due would have caused an 80 m eustatic change, assuming full to decreased residence time. compensation. On the myr scale considered here, compensa- tion was likely partially achieved (Peltier, 1997) and thus the Following the earliest Oligocene event, Oligocene to actual changes in water volume would have been higher than middle Miocene ice sheets vacillated from near modern vol- eustatic estimates. umes to nearly fully deglaciated at times. This was a transi- 5. The link of sea-level falls and 18O increases has tional period, with a wet-based Antarctic ice sheet (Marchant been documented for numerous icehouse increases. This et al., 1993). Atmospheric CO2 was at near-preanthropogenic relationship is expected because glacioeustasy is expected levels (Pagani et al., 2005). A ~1.2-1.5â° middle Miocene to drive sea-level change during this time. Further studies (ca. 14.8-12 Ma) 18O increase is associated with a major evaluating the link of 18O and sea level in the greenhouse sea-level fall, which this event most likely represents the are needed, though preliminary comparisons are intriguing. development of polar desert conditions with a permanent ice It must be remembered that we predict small (~0.2-0.3â°) 18 cap in Antarctica (Shackleton and Kennett, 1975) (Figure 2). Oseawater changes for greenhouse ice-volume changes. There has been intense debate about whether this ice cap was indeed permanent, or whether it in fact disintegrated in the Oxygen isotopic variations can be used to place con- early Pliocene (Kennett and Hodell, 1996). We note that the straints on Antarctic ice history. The ï¬rmest constraint is amplitude of the myr-scale variability in both 18O and sea- that ice sheets are required when benthic foraminiferal 18O level records appears to be lower after the middle Miocene values exceed 1.8â° in Cibicidoides (Miller and Fairbanks, event (Figure 2), suggesting relative stability. 1983; Miller et al., 1991). For the greenhouse world, 18O values suggest largely warm high latitudes, at least near the coast, where deep waters form. However, we need to be QUO VADIM? careful in assuming that warm coastal Antarctic tempera- We present a review of Antarctic glacial history as interpreted tures indicate a continent devoid of ice sheets for the period from sea-level and oxygen isotopic records, updating and 100-33.55 Ma. As models illustrate (DeConto and Pollard,
MILLER ET AL. 67 2003a,b), the Antarctic continent may have been remark- Barrett, P. J., D. P. Elston, D. M. Harwood, B. C. McKelvey, and P.-N. Webb. ably diverse during the greenhouse with unglaciated coastal 1987. Mid-Cenozoic record of glaciation and sea level change on the margin of the Victoria Land Basin, Antarctica. Geology 15:634-637. regions and small- to moderate-size ice sheets in the interior. Barron, J. A., and B. Larsen. 1989. Proceedings of the Ocean Drilling Though much of the evidence may have been destroyed or Program, Initial Reports, vol. 119. College Station, TX: Ocean Drill- buried by the modern ice sheet, the challenge now is to seek ing Program. evidence for these greenhouse ice sheets. Intense study of the Berggren, W. A., D. V. Kent, C. C. Swisher, and M.-P. Aubry. 1995. A revised Antarctic continent has demonstrated the importance of ice Cenozoic geochronology and chronostratigraphy. In Geochronology, Time Scales and Global Stratigraphic Correlations: A Uniï¬ed Temporal sheets on sedimentation over the past 33 myr and provided Framework for an Historical Geology, eds. W. A. Berggren, D. V. Kent, hints of cold conditions even in the Eocene (Figure 1). Com- M.-P. Aubry, and J. Hardenbol, pp. 129-212. Special Publication. Tulsa, parison with the Arctic is illustrative. Until the most sensitive OK: Society for Sedimentary Geology. area of the Arctic was drilled by IODP (Moran et al., 2006), it Birkenmajer, K. 1991. Tertiary glaciation in the South Shetland Islands, was assumed that the Arctic was not glaciated until the later West Antarctica: Evaluation of data. In Geological Evolution of Ant- arctica, eds. M. R. A. Thomson, J. A. Crame, and J. W. Thomson, pp. Cenozoic and bipolar glaciation was a relatively recent event. 629-632. Cambridge: Cambridge University Press. In contrast, sampling on the Lomonsov Ridge has extended Birkenmajer, K., A. Gazdzicki, K. P. Krajewski, A. Przybycin, A. Solecki, A. the glacial record of the Arctic back to ~46 Ma (Moran et al., Tatur, and H. I. Oon. 2005. First Cenozoic glaciers in West Antarctica. 2006), and we predict that future studies of Antarctica will Polish Polar Research 26:3-12. result in extension of small- to medium-size ice sheets back Breza, J. R., and S. W. Wise, Jr. 1992. Lower Oligocene ice-rafted debris on the Kerguelen Plateau: Evidence for East Antarctic continental glacia- into the Late Cretaceous. tion. In Proceedings of the Ocean Drilling Program, Scientiï¬c Results, vol. 120, eds. S. W. Wise, Jr. and R. Schlich, pp. 161-178. College Sta- tion, TX: Ocean Drilling Program. ACKNOWLEDGMENTS Browning, J. V., K. G. Miller, and D. K. Pak. 1996. Global implications of We thank J. P. Kennett and S. F. Pekar for reviews and A. lower to middle Eocene sequences on the New Jersey coastal plain: The icehouse cometh. Geology 24:639-642. Cooper for suggestions. The ideas presented here were devel- Browning, J. V., K. G. Miller, P. P. McLaughlin, M. A. Kominz, P. J. oped with help from previous reviews of sea-level change Sugarman, D. Monteverde, M. D. Feigenson, and J. C. HernÃ ndez. 2006. (Kominz et al., in review; Miller et al., 2005a), greenhouse Quantiï¬cation of the effects of eustasy, subsidence, and sediment sup- ice sheets (Miller et al., 2005a), the Eocene-Oligocene transi- ply on Miocene sequences, Mid-Atlantic margin of the United States. tion (Miller et al., 2008), and deep-sea circulation changes Geological Society of America Bulletin 118:567-588. Christie-Blick, N., G. S. Mountain, and K. G. Miller. 1990. Seismic strati- (Wright and Miller, 1996); we thank co-authors of those graphic record of sea-level change. In Sea-Level Change, pp. 116-140. papers, who are not listed here, for helping to develop our Washington, D.C.: National Academy Press. views of Antarctic glacial history. Supported by National Sci- Cooper, A. K., and P. E. OâBrien. 2004. Leg 188 synthesis: Transitions ence Foundation grants EAR06-06693 (Miller) and OCE06- in the glacial history of the Prydz Bay region, East Antarctica, from 23256 (Katz, Miller, Wade, and Wright). ODP drilling. In Proceedings of the Ocean Drilling Program, Scientiï¬c Results, vol. 188, eds. A. K. Cooper, P. E. OâBrien, and C. Richter, pp. 1-42. College Station, TX: Ocean Drilling Program. REFERENCES Cooper, A. K., G. Brancolini, E. Escutia, Y. Kristoffersen, R. Larter, G. Leitchenkov, P. OâBrien, and W. Jokat. Forthcoming. Cenozoic climate Ali, J., and E. A. Hailwood. 1995. Magnetostratigraphy of upper Paleocene history from seismic-reï¬ection and drilling studies on the Antarctic con- through lower middle Eocene strata of northwest Europe. In Geochro- tinental margin. In Antarctic Climate Evolution. Amsterdam: Elsevier. nology, Time Scales and Global Stratigraphic Correlation, eds. W. A. Coxall, H. K., P. A. Wilson, H. PÃ¤like, C. H. Lear, and J. Backman. 2005. Berggren, D. V. Kent, M.-P. Aubry, and J. Hardenbol, pp. 275-279. Rapid stepwise onset of Antarctic glaciation and deeper calcite compen- Special Publication. Tulsa, OK: Society for Sedimentary Geology. sation in the Paciï¬c Ocean. Nature 433:53-57. Askin, R. A. 1989. Endemism and heterochroneity in the Late Cretaceous DeConto, R. M., and D. Pollard. 2003a. A coupled climate-ice sheet model- (Campanian) to Paleocene palynoï¬oras of Seymour Island, Antarctica: ing approach to the early Cenozoic history of the Antarctic ice sheet. Implications for origins, dispersal and palaeoclimates of southern ï¬oras. Paleogeography, Paleoclimatology, Paleoecology 198:39-52. In Origins and Evolution of the Antarctic Biota, ed. J. A. Crame, pp. DeConto, R. M., and D. Pollard. 2003b. Rapid Cenozoic glaciation of Ant- 107-119. Special Publication. Geological Society of London. arctica induced by declining atmospheric CO2. Nature 421:245-249. Barrera, E., and B. T. Huber. 1990. Evolution of Antarctic waters during the Diester-Haass, L., and J. Zachos. 2003. The Eocene-Oligocene transition in Maestrichtian: foraminifer oxygen and carbon isotope ratios, ODP Leg the Equatorial Atlantic (ODP Site 925); paleoproductivity increase and 113. In Proceedings of the Ocean Drilling Program, Scientiï¬c Results, positive 13C excursion. In Greenhouse to Icehouse: The Marine Eocene- vol. 113, eds. P. F. Barker and J. P. Kennett, pp. 813-823. College Station, Oligocene Transition, eds. D. R. Prothero, L. C. Ivany, and E. A. Nesbitt. TX: Ocean Drilling Program. pp. 397-416, New York: Columbia University Press. Barrera, E., and S. M. Savin. 1999. Evolution of late Campanian-Maastrich- Eldrett, J. S., I. C. Harding, P. A. Wilson, E. Butler, and A. P. Roberts. tian marine climates and oceans. Geological Society of America Special 2007. Continental ice in Greenland during the Eocene and Oligocene. Papers 332:245-282. Science 446:176-179. Barrett, P. J. 2007. Cenozoic climate and sea level history from glacimarine Emiliani, C. 1955. Pleistocene temperatures. Journal of Geology 63: strata off the Victoria Land coast, Cape Roberts Project, Antarctica. In 538-578. Glacial Processes and Products, eds. M. J. Hambrey, P. Christoffersen, Fairbanks, R. G. 1989. A 17,000-year glacio-eustatic sea level record: Inï¬u- N. F. Glasser, and B. Hubbart. International Association of Sedimentolo- ence of glacial melting rates on the Younger Dryas event and deep-ocean gists Special Publication 39:259-287. circulation. Nature 342:637-642.
68 ANTARCTICA: A KEYSTONE IN A CHANGING WORLD Falkowski, P. G., M. E. Katz, A. Knoll, A. Quigg, J. A. Raven, O. Schoï¬eld, Leckie, M., and P.-N. Webb. 1986. Late Paleogene and early Neogene and M. Taylor. 2004. The evolutionary history of eukaryotic phytoplank- foraminifers of Deep Sea Drilling Project Site 270, Ross Sea, Initial ton. Science 305:354-360. Reports DSDP 90:1093-1142. Washington, D.C.: U.S. Government Finkel, Z. V., M. E. Katz, J. D. Wright, O. M. E. Schoï¬eld, and P. G. Printing Ofï¬ce. Falkowski. 2005. Climatically-driven evolutionary change in the size LeMasurier, W. E., and D. C. Rex. 1982. Volcanic record of Cenozoic glacial of diatoms over the Cenozoic. Proceedings of the National Academy of history in Marie Byrd Land and western Ellsworth Land: Revised chro- Sciences U.S.A. 102:8927-8932. nology and evaluation of tectonic factors. In Antarctic Geoscience, ed. Fitzgerald, P. G. 2002. Tectonics and landscape evolution of the Ant- C. Craddock, pp. 725-732. Madison: University of Wisconsin Press. arctic plate since Gondwana breakup, with an emphasis on the West Mancini, E. A., and B. H. Tew. 1995. Geochronology, biostratigraphy and Antarctic rift system and the Transantarctic Mountains. In Antarctica sequence stratigraphy of a marginal marine to marine shelf stratigraphic at the Close of a Millennium. Proceedings of the 8th International succession: Upper Paleocene and lower Eocene, Wilcox Group, eastern Symposium on Antarctic Earth Science, eds. J. A. Gamble, D. N. B. Gulf Coastal Plain, U.S.A. In Geochronology, Time Scales and Global Skinner, and S. Henrys. Bulletin of the Royal Society of New Zealand Stratigraphic Correlations: A Uniï¬ed Temporal Framework for an 35:453-469. Historical Geology, eds. W. A. Berggren, D. V. Kent, M.-P. Aubry, and Flint, R. F. 1971. Glacial and Quaternary Geology. New York: John J. Hardenbol, pp. 281-293. Special Publication 54. Tulsa, OK: Society Wiley. for Sedimentary Geology. Francis, J. E., and I. Poole. 2002. Cretaceous and early Tertiary climates of Marchant, D. R., G. H. Denton, and C. C. Swisher. 1993. Miocene-Pliocene- Antarctica: Evidence from fossil wood. Paleogeography, Paleoclimatol- Pleistocene glacial history of Arena Valley, Quartermain Mountains, ogy, Paleoecology 182:47-64. Antarctica. Geograï¬ska Annaler 75A:269-302. Hancock, J. M. 1993. Transatlantic correlations in the Campanian-Maas- Margolis, S. V., and J. P. Kennett. 1971. Cenozoic paleoglacial history of trichtian stages by eustatic changes of sea-level. In High Resolution Antarctica recorded in subantarctic deep-sea cores. American Journal Stratigraphy, eds. E. A. Hailwood and R. B. Kidd, pp. 241-256. Geologi- of Science 271:1-36. cal Society Special Publication. Matthews, R. K. 1984. Oxygen-isotope record of ice-volume history: 100 Haq, B. U., J. Hardenbol, and P. R. Vail. 1987. Chronology of ï¬uctuating million years of glacio-eustatic sea-level ï¬uctuation. Memoirs of the sea levels since the Triassic (250 million years ago to present). Science American Association of Petroleum Geologists 36:97-107. 235:1156-1167. Matthews, R. K., and C. Frohlich. 2002. Maximum ï¬ooding surfaces Hays, J. D., J. Imbrie, and N. J. Shackleton. 1976. Variations in the earthâs and sequence boundaries: Comparisons between observations and orbit: Pacemaker of the ice ages. Science 194:1121-1132. orbital forcing in the Cretaceous and Jurassic (65-190 Ma). GeoArabia Hedberg, H. D. 1976. International Stratigraphic Guide. New York: 7:503-538. Wiley-Interscience. Matthews, R. K., and R. Z. Poore. 1980. Tertiary 18O record and glacio- Huber, B. T., R. D. Norris, and K. G. MacLeod. 2002. Deep sea paleotem- eustatic sea-level ï¬uctuations. Geology 8:501-504. perature record of extreme warmth during the Cretaceous. Geology Miall, A. D. 1991. Stratigraphic sequences and their chronostratigraphic 30:123-126. correlation. Journal of Sedimentary Petrology 61:497-505. Ivany, C. L., S. Van Simaeys, E. W. Domack, and S. D. Samson. 2006. Miller, K. G. 1992. Middle Eocene to Oligocene stable isotopes, climate, and Evidence for an earliest Oligocene ice sheet on the Antarctic Peninsula. deep-water history: The Terminal Eocene Event? In Eocene-Oligocene Geology 34:377-380. Climatic and Biotic Evolution, eds. D. Prothero and W. A. Berggren, pp. Kennett, J. P. 1977. Cenozoic evolution of Antarctic glaciation, the Circum- 160-177. Princeton: Princeton University Press. Antarctic Ocean, and their impact on global paleoceanography. Journal Miller, K. G., and R. G. Fairbanks. 1983. Evidence for Oligocene-middle of Geophysical Research 82:3843-3860. Miocene abyssal circulation changes in the western North Atlantic. Kennett, J. P., and P. F. Barker. 1990. Latest Cretaceous to Cenozoic climate Nature 306:250-252. and oceanographic developments in the Weddell Sea, Antarctica: An Miller, K. G., and R. G. Fairbanks. 1985. Oligocene to Miocene carbon ocean-drilling perspective. In Proceedings of the Ocean Drilling Pro- isotope cycles and abyssal circulation changes. In The Carbon Cycle gram, Scientiï¬c Results, vol. 113, eds. P. F. Barker and J. P. Kennett, pp. and Atmospheric CO2: Natural Variations Archean to Present, eds. 937-960. College Station, TX: Ocean Drilling Program. E. T. Sundquist and W. S. Broecker, pp. 469-486. Washington, D.C.: Kennett, J. P., and D. A. Hodell. 1996. Stability or instability of Antarctic ice American Geophysical Union. sheets during warm climates of the Pliocene? GSA Today 5:1-22. Miller, K. G., and M. E. Katz. 1987. Oligocene-Miocene benthic forami- Kennett, J. P., and N. J. Shackleton. 1976. Oxygen isotopic evidence niferal and abyssal circulation changes in the North Atlantic. Micropa- for the development of the psychrosphere 38 Myr ago. Nature leontology 33:97-149. 260:513-515. Miller, K. G., and B. E. Tucholke. 1983. Development of Cenozoic abyssal cir- Kennett, J., and L. Stott. 1990. Proteus and Proto-Oceanus: Ancestral culation south of the Greenland-Scotland Ridge. In Structure and Devel- Paleogene oceans as revealed from Antarctic stable isotopic results, opment of the Greenland-Scotland Ridge, eds. M. H. P. Bott, S. Saxov, M. ODP Leg 113. In Proceedings ODP Scientiï¬c Results, vol. 113, eds. Talwani, and J. Thiede, pp. 549-589. New York: Plenum Press. P. Barker and J. Kennett, pp. 865-880. College Station, TX: Ocean Miller, K. G., R. G. Fairbanks, and G. S. Mountain. 1987. Tertiary oxygen Drilling Program. isotope synthesis, sea level history, and continental margin erosion. Kominz, M. A., K. G. Miller, and J. V. Browning. 1998. Long-term and Paleoceanography 2:1-19. short-term global Cenozoic sea-level estimates. Geology 26:311-314. Miller, K. G., J. D. Wright, and R. G. Fairbanks. 1991. Unlocking the Ice Kominz, M. A., J. V. Browning, K. G. Miller, P. J. Sugarman, S. Mizintseva, House: Oligocene-Miocene oxygen isotopes, eustasy, and margin ero- A. Harris, and C. R. Scotese. In review. Late Cretaceous to Miocene sion. Journal of Geophysical Research 96:6829-6848. sea-level estimates from the New Jersey and Delaware coastal plain Miller, K. G., G. S. Mountain, the Leg 150 Shipboard Party, and Members coreholes: An error analysis. Basin Research. of the New Jersey Coastal Plain Drilling Project. 1996. Drilling and Larsen, H. C., A. D. Saunders, P. D. Clift, J. Beget, W. Wei, S. Spezzaferri, dating New Jersey Oligocene-Miocene sequences: Ice volume, global and ODP Leg 152 Scientiï¬c Party. 1994. Seven million years of glacia- sea level, and Exxon records. Science 271:1092-1094. tion in Greenland. Science 264:952-955. Miller, K. G., G. S. Mountain, J. V. Browning, M. A. Kominz, P. J. Sugar- Lear, C. H., and Y. Rosenthal. 2006. Benthic foraminiferal Li/Ca: Insights into man, N. Christie-Blick, M. E. Katz, and J. D. Wright. 1998. Cenozoic Cenozoic seawater carbonate saturation state. Geology 34:985-988. global sea-level, sequences, and the New Jersey transect: Results from coastal plain and slope drilling. Reviews of Geophysics 36:569-601.
MILLER ET AL. 69 Miller, K. G., E. Barrera, R. K. Olsson, P. J. Sugarman, and S. M. Royer, D., R. A. Berner, I. P. Montanez, N. J. Tabor, and D. J. Beerling. 2004. Savin. 1999. Does ice drive early Maastrichtian eustasy? Geology CO2 as a primary driver of Phanerozoic climate. GSA Today 14:4-10. 27:783-786. Sahagian, D., O. Pinous, A. Olferiev, V. Zakaharov, and A. Beisel. 1996. Miller, K. G., P. J. Sugarman, J. V. Browning, M. A. Kominz, J. C. HernÃ ndez, Eustatic curve for the Middle Jurassic-Cretaceous based on Russian R. K. Olsson, J. D. Wright, M. D. Feigenson, and W. Van Sickel. 2003. platform and Siberian stratigraphy: Zonal resolution. American Associa- Late Cretaceous chronology of large, rapid sea-level changes: Gla- tion of Petroleum Geologists Bulletin 80:1433-1458. cioeustasy during the greenhouse world. Geology 31:585-588. Salamy, K. A., and J. C. Zachos. 1999. Late Eocene-Early Oligocene Miller, K. G., P. J. Sugarman, J. V. Browning, M. A. Kominz, R. K. Olsson, climate change on southern ocean fertility: Inferences from sediment M. D. Feigenson, and J. C. HernÃ ndez. 2004. Upper Cretaceous accumulation and stable isotope data. Paleogeography, Paleoclimatol- sequences and sea-level history, New Jersey coastal plain. Geological ogy, Paleoecology 145:79-93. Society of America Bulletin 116:368-393. Savin, S. M., R. G. Douglas, and F. G. Stehli. 1975. Tertiary marine paleo- Miller, K. G., M. A. Kominz, J. V. Browning, J. D. Wright, G. S. Mountain, temperatures. Geological Society of America Bulletin 86:1499-1510. M. E. Katz, P. J. Sugarman, B. S. Cramer, N. Christie-Blick, and S. F. Shackleton, N. J. 1967. Oxygen isotope analyses and Pleistocene tempera- Pekar. 2005a. The Phanerozoic record of global sea-level change. Sci- tures re-assessed. Nature 215:15-17. ence 310:1293-1298. Shackleton, N. J., and J. P. Kennett. 1975. Paleotemperature history of the Miller, K. G., J. D. Wright, and J. V. Browning. 2005b. Visions of ice sheets Cenozoic and the initiation of Antarctic glaciation, oxygen and carbon in a greenhouse world. Marine Geology 217:215-231. isotope analyses in DSDP sites 277, 279, and 281. In Init. Reports DSDP, Miller, K. G., J. V. Browning, M.-P. Aubry, B. S. Wade, M. E. Katz, A. A. eds. J. P. Kennett, R. E. Houtz et al., pp. 743-755. Washington, D.C.: Kulpecz, and J. D. Wright. 2008. Eocene-Oligocene global climate and U.S. Government Printing Ofï¬ce. sea-level changes: St. Stephens Quarry, Alabama. Geological Society of Shackleton, N. J., J. Backman, H. Zimmerman, D. V. Kent, M. A. Hall, D. G. America Bulletin 120:34-53. Roberts, D. Schnitker, J. G. Baldauf, A. Desprairies, R. Homrighausen, Moore, T. C., Jr., J. Backman, I. Rafï¬, C. Nigrini, A. Sanï¬lippo, H. P. Huddlestun, J. B. Keene, A. J. Kaltenback, K. A. O. Krumsiek, A. C. PÃ¤like, and M. Lyle. 2004. Paleogene tropical Pacific: Clues to Morton, J. W. Murray, and J. Westberg-Smith. 1984. Oxygen isotope circulation, productivity and plate motion. Paleoceanography 19, calibration of the onset of ice-rafting and history of glaciation in the doi:10.1029/2003PA000998. North Atlantic region. Nature 307:620-623. Moran, K., J. Backman, H. Brinkhuis, S. C. Clemens, T. Cronin, Stoll, H. M., and D. P. Schrag. 1996. Evidence for glacial control of rapid G. R. Dickens, F. Eynaud, J. Gattacceca, M. Jakobsson, R. W. Jordan, sea level changes in the Early Cretaceous. Science 272:1771-1774. M. Kaminski, J. King, N. Koc, A. Krylov, N. Martinez, J. Matthiessen, Stoll, H. M., and D. P. Schrag. 2000. High resolution stable isotope records D. McInroy, T. M. Moore, J. Onodera, M. OâRegan, H. Palike, from the Upper Cretaceous rocks of Italy and Spain: Glacial episodes B. Rea, D. Rio, T. Sakamoto, D. C. Smith, R. Stein, K. St John, in a greenhouse planet? Geological Society of America Bulletin I. Suto, N. Suzuki, K. Takahashi, M. Watanabe, M. Yamamoto, J. Farrell, 112:308-319. M. Frank, P. Kubik, W. Jokat, and Y. Kristoffersen. 2006. The Cenozoic Strand, K., S. Passchier, and J. Nasi. 2003. Implications of quartz grain palaeoenvironment of the Arctic Ocean. Nature 441:601-605. microtextures for onset Eocene/Oligocene glaciation in Prydz Bay, ODP Moriya, K., P. A. Wilson, O. Friedrich, J. Erbacher, and H. Kawahata. Site 1166, Antarctica. Paleogeography, Paleoclimatology, Paleoecology 2007. Testing for ice sheets during the mid-Cretaceous greenhouse 198:101-111. using glassy foraminiferal calcite from the mid-Cenomanian tropics on Suko, I. 2006. The explosive diversiï¬cation of the diatom genus Chaetoceros Demerara Rise. Geology 35:615-618. across the Eocene/Oligocene and Oligocene/Miocene boundaries in the Mountain, G. S., and K. G. Miller. 1992. Seismic and geologic evidence for Norwegian Sea. Marine Micropaleontology 58:259-269. early Paleogene deep-water circulation in the western North Atlantic. ten Brink, U., R. Hackney, S. Bannister, T. Stern, and Y. Makovsky. Paleoceanography 7:423-439. 1997. Uplift of the Transantarctic Mountains and the bedrock beneath Pagani, M., J. Zachos, K. H. Freeman, S. Bohaty, and B. Tipple. 2005. the East Antarctic ice sheet. Journal of Geophysical Research 102: Marked change in atmospheric carbon dioxide concentrations during 27603-27621. the Oligocene. Science 309:600-603. Thomas, E., ed. 1990. Late Cretaceous through Neogene Deep-Sea Benthic Pak, D. K., and K. G. Miller. 1992. Paleocene to Eocene benthic foraminif- Foraminifers (Maud Rise, Weddell Sea, Antarctica). College Station, TX: eral isotopes and assemblages: Implications for deepwater circulation. Ocean Drilling Program. Paleoceanography 7:405-422. Thorne, J. A., and A. B. Watts. 1984. Seismic reï¬ectors and unconformities Pekar, S., and K. G. Miller. 1996. New Jersey Oligocene âIcehouseâ at passive continental margins. Nature 311:365-368. sequences (ODP Leg 150X) correlated with global 18O and Exxon Thunell, R. C., and B. H. Corliss. 1986. Late Eocene-early Oligocene car- eustatic records. Geology 24:567-570. bonate sedimentation in the deep sea. In Terminal Eocene Events, eds. C. Pekar, S. F., N. Christie-Blick, M. A. Kominz, and K. G. Miller. 2002. Pomerol and I. Premoli-Silva, pp. 363-380. Amsterdam: Elsevier. Calibration between eustatic estimates from backstripping and oxygen Tripati, A., J. Backman, H. Elderï¬eld, and P. Ferretti. 2005. Eocene bipo- isotopic records for the Oligocene. Geology 30:903-906. lar glaciation associated with global carbon cycle changes. Nature Peltier, W. R. 1997. Postglacial variations in the level of the sea: Implications 436:341-346. for climate dynamics and solid-earth geophysics. Reviews of Geophys- Troedson, A. L., and J. B. Riding. 2002. Upper Oligocene to lowermost Mio- ics 36:603-689. cene strata of King George Island, South Shetland Islands, Antarctica: Pitman, W. C., III, and X. Golovchenko. 1983. The effect of sea-level change Stratigraphy, facies analysis and implications for the glacial history of on the shelf edge and slope of passive margins. Special Publication 33, the Antarctic Peninsula. Journal of Sedimentary Research 72:510-523. pp. 41-58. Society of Economic Paleontologists and Mineralogists. Troedson, A. L., and J. L. Smellie. 2002. The Polonez Cove Formation of Rea, D. K., and M. W. Lyle. 2005. Paleogene calcite compensation depth in King George Island, Antarctica: Stratigraphy, facies and implications for the eastern 340 subtropical Paciï¬c; answers and questions. Paleocean- mid-Cenozoic cryosphere development. Sedimentology 49:277-301. ography 20:1-9. Tucholke, B. E., and G. S. Mountain. 1979. Seismic stratigraphy, lithostratig- Robert, C., and J. P. Kennett. 1997. Antarctic continental weathering raphy and paleosedimentation patterns in the North American basin. In changes during Eocene-Oligocene cryosphere expansion: Clay mineral Deep Drilling Results in the Atlantic Ocean: Continental Margins and and oxygen isotope evidence Geology 25:587-590. Paleoenvironment, eds. M. W. Talwani, W. Hay, and W. B. F. Ryan, pp. 58-86. Washington, D.C.: American Geophysical Union.
70 ANTARCTICA: A KEYSTONE IN A CHANGING WORLD Vail, P. R., R. M. Mitchum, R. G. Todd, J. M. Widmier, S. Thompson III, J. Wise, S. W., Jr., J. R. Breza, D. M. Harwood, and W. Wei. 1991. Paleogene B. Sangree, J. N. Bubb, and W. G. Hatlelid. 1977. Seismic stratigraphy glacial history of Antarctica. In Controversies in Modern Geology: and global changes of sea level. In Seismic StratigraphyâApplications Evolution of Geological Theories in Sedimentology, Earth History and to Hydrocarbon Exploration, ed. C. E. Payton. Memoirs of the American Tectonics, eds. D. W. Miller, J. A. McKenzie, and H. Weissert, pp. 133- Association of Petroleum Geologists 26:49-205. Tulsa, OK: AAPG. 171. London: Academic Press. van Andel, T. H. 1975. Mesozoic/Cenozoic calcite compensation depth and Wright, J. D., and K. G. Miller. 1996. Control of North Atlantic deep global distribution of calcareous sediments. Earth and Planetary Science water circulation by the Greenland-Scotland Ridge. Paleoceanography Letters 26:187-194. 11:157-170. Van Sickel, W. A., M. A. Kominz, K. G. Miller, and J. V. Browning. 2004. Zachos, J. C., J. Breza, and S. W. Wise. 1992. Earliest Oligocene ice-sheet Late Cretaceous and Cenozoic sea-level estimates: Backstripping analy- expansion on East Antarctica: Stable isotope and sedimentological data sis of borehole data, onshore New Jersey. Basin Research 16:451-465. from Kerguelen Plateau. Geology 20:569-573. Wade, B. S., and H. PÃ¤like. 2004. Oligocene climate dynamics. Paleocean- Zachos, J. C., L. D. Stott, and K. C. Lohmann. 1994. Evolution of early ography 19:PA4019. doi:10.1029/2004PA001042. Cenozoic marine temperatures. Paleoceanography 9:353-387. Wei, W. 1992. Calcareous nannofossil stratigraphy and reassessment of Zachos, J. C., T. M. Quinn, and K. A. Salamy. 1996. High resolution (104 yr) the Eocene glacial record in subantarctic piston cores of the southeast deep-sea foraminiferal stable isotope records of the earliest Oligocene Paciï¬c. In Proceedings of the ODP Scientiï¬c Results, eds. S. W. Wise, climate transition. Paleoceanography 9:353-387. Jr. and R. Schlich, pp. 1093-1104. College Station, TX: Ocean Drilling Program. Williams, R. S., and J. G. Ferrigno. 1995. Satellite image atlas of glaciers of the worldâGreenland. U.S. Geological Survey Professional Paper 1386-C: 141 pp, http://pubs.usgs.gov/pp/p1386c/p1386c.pdf.