13C and alkalinity, point toward greenhouse forcing (Shackleton, 1986; Kennett and Stott, 1991; Zachos et al., 1993; Thomas and Shackleton, 1996). Samples recovered during Leg 198 will help constrain the nature and causes of these warm episodes.
The exact causes of the long- and short-term Paleogene warm episodes remain enigmatic. Several pieces of geochemical evidence, including changes in the mean ocean 13C and alkalinity, point toward greenhouse forcing (Shackleton, 1986; Kennett and Stott, 1991; Zachos et al., 1993; Thomas and Shackleton, 1996). Samples recovered during Leg 198 will help constrain the nature and causes of these warm episodes.
In terms of the rate and degree of warming, the PETM is unprecedented in Earth history (Fig. F6). The deep-sea and high-latitude oceans warmed by 4°C and 8°C, respectively. The carbon isotopic composition of the ocean decreased by 3
-4 coeval with the warming event, suggesting a massive perturbation to the global carbon cycle (Fig. F7) (Kennett and Stott, 1991; Bains et al., 1999). The large magnitude and rate (~-3-4/5 k.y.) of the carbon isotope excursion (CIE) is consistent with the sudden injection of a large volume of methane from clathrates stored in continental slope sediments (Dickens et al., 1995, 1997). Much of this methane would have quickly converted to CO2, stripping O2 from deep waters, contributing to the major extinction event of benthic foraminifers (Thomas, 1990), and lowering alkalinity. The result should be a sharp rise in the lysocline and CCD (Dickens, 2000). Both CO2 and CH4 would also have immediately contributed to greenhouse warming.
The Leg 198 depth transect will help us determine (1) the magnitude of the tropical Pacific sea-surface and deepwater temperature increase during the PETM; (2) whether or not the Pacific lysocline and CCD shoaled during the CIE, whether or not bottom-water oxygenation decreased, and how these changes fit with geochemical models of clathrate release; (3) the response of planktonic and benthic populations to the PETM in the subtropical Pacific; and (4) whether or not there is a change in the distribution of bottom-water carbon isotopes prior to and/or during the PETM signaling possible circulation changes.
Several investigators have suggested that early Cenozoic global warming would have altered deep-ocean circulation patterns by reducing the density of surface waters at high latitudes (Kennett and Shackleton, 1976; Wright and Miller, 1993; Zachos et al., 1993). This, in turn, would permit increased downwelling of highly saline but warmer waters in subtropical oceans. Such reversals or switches in circulation probably occurred suddenly rather than gradually. In fact, it has been suggested that a sudden change in intermediate-water circulation patterns may have occurred just prior to the PETM, possibly triggering the dissociation of clathrates (Bralower et al., 1997a). There may have been additional, abrupt warming intervals in the late Paleocene and early Eocene (Thomas and Zachos, 1999; Thomas et al., 2000). These "hyperthermals" were characterized by changes in the assemblage composition of benthic foraminifers corresponding to negative shifts in planktonic and benthic foraminiferal 18O and 13C values. The ultimate cause of the hyperthermals may be similar to the PETM, driven by the release of greenhouse gas.
Leg 198 samples will be used to assess regional and global circulation changes during the Paleogene. Major changes in the sources of waters bathing Shatsky Rise might be reflected in the spatial and vertical distribution of carbon isotope ratios in bottom waters as well as in benthic foraminiferal assemblage patterns. Several studies have shown that throughout the late Paleocene and early Eocene, the most negative deep-ocean carbon isotope values were consistently recorded by benthic foraminifers from Shatsky Rise (Miller et al., 1987b; Pak and Miller, 1992; Corfield et al., 1991). Such a pattern is similar to that in the modern ocean, implying older, nutrient-enriched waters in the Pacific, and younger, nutrient-depleted waters in the high latitudes. Although Site 577 is discontinuous across the Paleocene/Eocene (P/E) boundary, isotope data from Site 865 on Allison Guyot in the equatorial Pacific suggest a possible reduction, if not reversal, in the 13C gradient between the shallow Pacific and the rest of the ocean (Bralower et al., 1995). If this was true, it would be consistent with increased production of intermediate waters in low latitudes. In summary, Leg 198 samples will help address whether there is evidence of warmer, more saline deep waters at times during the Paleogene and how export production in the Pacific changed from the Paleocene to the Eocene.
The Eocene-Oligocene represents the final transition from a "greenhouse" to an "icehouse" world. Although this transition occurred over a period of 18 m.y., stable isotopic records reveal that much of the cooling occurred over relatively brief intervals in the late early Eocene (~50-51 Ma) and earliest Oligocene (~33 Ma) (Fig. F6) (e.g., Kennett, 1977; Miller et al., 1987a; Stott et al., 1990; Miller et al., 1991; Zachos et al., 1996). Furthermore, small, ephemeral ice sheets were probably present on Antarctica sometime after the first event (Browning et al., 1996). The first large permanent ice sheets became established much later, most likely during the early Oligocene event (Zachos et al., 1992a). Current reconstructions of ocean temperature and chemistry for the Eocene and Oligocene, however, are based primarily on pelagic sediments collected in the Atlantic and Indian Oceans (Miller et al., 1987a; Zachos et al., 1992b, 1996). Very few sections suitable for such work have been recovered from the Pacific (Miller and Thomas, 1985; Miller and Fairbanks, 1985). As a consequence, we still lack a robust understanding of how global ocean chemistry and circulation evolved in response to high-latitude cooling and glaciation.
Leg 198 sections across the Eocene-Oligocene transition will provide a vertical depth transect of ocean chemistry and temperature changes during this important climatic transition. These sections will allow us to determine whether the basin-to-basin deep carbon isotope gradient changed during the Eocene-Oligocene transition in response to high-latitude cooling and glaciation, and how the lysocline/CCD in the Pacific responded to the rapid high-latitude cooling/glaciation.
The mid-Cretaceous (Barremian-Turonian [125-90 Ma]) Earth experienced some of the warmest temperatures and lowest thermal gradients of the entire Phanerozoic Eon. This time interval, therefore, represents one of the best ancient approximations of "greenhouse" climate. The Late Cretaceous was characterized by significant global cooling, but available oxygen isotopic records differ on the exact timing of the end of the "greenhouse" conditions. Records from DSDP Site 511 on the Falkland Plateau, South Atlantic (Huber et al., 1995), and the chalk from England (Jenkyns et al., 1994) suggest that peak warmth occurred in the early Turonian, at ~90 Ma (Fig. F8). Data from Shatsky Rise DSDP sites (e.g., Douglas and Savin, 1975; Savin 1977), however, indicate that peak "greenhouse" conditions existed in the Albian, at ~105 Ma. In addition, these stratigraphies differ on whether peak warming was immediately followed by long-term cooling (English chalk) or sustained warmth then cooling beginning in the mid-Campanian (Site 511 data). Differences between the various records may reflect real latitudinal climatic variations or diagenetic alteration of stable isotopic proxies.
There is also significant disparity as to exactly how much cooling occurred in the Late Cretaceous, especially in the tropics. Savin (1977) and D'Hondt and Arthur (1996) concluded that the Maastrichtian was characterized by surprisingly cool tropical SSTs (20°-21°C) based on 18O analyses of planktonic foraminifers, the "cool tropics paradox" (D'Hondt and Arthur, 1996). Wilson and Opdyke (1996), on the other hand, measured 18O values on rudists recovered from Pacific guyots and concluded that tropical SSTs in the same interval were extremely warm (between 27° and 32°C). More recently, Pearson et al. (2001) have also found evidence for high Maastrichtian temperatures in planktonic foraminifers from clay units in Tanzania. The climate history of the Cretaceous is based on a limited number of oxygen isotope data points from few sites with little information from the tropics. In fact, Shatsky Rise Site 305 (Douglas and Savin, 1975) is among a handful of low-latitude sites that form the basis of most Cretaceous thermal gradient estimates that are used as inputs in climate models (e.g., Barron and Peterson, 1991).
There is a limited understanding of the evolution of bottom-water circulation in the mid- and Late Cretaceous, in particular, of how and when the transition from low-latitude (e.g., Brass et al., 1982) to high-latitude (e.g., Zachos et al., 1993) deepwater sources took place. Benthic foraminiferal 18O records are even sparser than those based on planktonic foraminifers, and there are very few benthic data from the entire Pacific. Thus, the role of this giant basin in the evolution of deep waters during the mid- and Late Cretaceous is poorly understood.
The long-term cooling of the Late Cretaceous was interrupted by a significant event in the mid-Maastrichtian, when the source of deep waters appears to have changed abruptly from low- to high-latitude sources (e.g., MacLeod and Huber, 1996; Barrera et al., 1997; Frank and Arthur, 1999). This event appears to have coincided with the extinction of the inoceramid bivalves (MacLeod et al., 1996) and possibly also the rudistid bivalves (Johnson et al., 1996). Growing evidence, however, suggests that this benthic event is distinctly diachronous (MacLeod et al., 1996). The change to high-latitude deepwater sources appears to have been long-lived, lasting until the PETM. However, more benthic data are required to accurately characterize Late Cretaceous and Paleocene deepwater properties.
The Upper Cretaceous sediments recovered during Leg 198 will help us determine (1) whether peak "greenhouse" conditions occurred in the Albian (Savin, 1977) or the early Turonian (Jenkyns et al., 1994; Huber et al., 1995); (2) if peak warming was immediately followed by long-term cooling (Jenkyns et al., 1994) or by sustained warmth and then by cooling beginning in the mid-Campanian (Huber et al., 1995), or whether cooling history varied between latitudes; (3) whether apparent cool tropical temperatures in the Maastrichtian (the "cool tropics paradox" of D'Hondt and Arthur [1996]) were real or the result of diagenetic alteration of planktonic foraminiferal tests; (4) the properties and age of mid- and Late Cretaceous deep water and from what oceanic region it was derived; (5) the timing and rate of changes in the sources of deep waters from low to high latitudes; (6) the changes in deepwater mass properties that accompanied the MME and their effect on benthic faunas; (7) whether the MME led to a permanent change in deepwater source; and (8) the changes in vertical thermal gradients through time and whether climate and deepwater circulational changes were coupled.
The beginning of "greenhouse" climate conditions in the mid-Cretaceous (Barremian-Turonian) was associated with widespread deposition of Corg-rich sediments, informally known as "black shales," in the oceans. These Corg-rich deposits are known to occur primarily at specific stratigraphic horizons, namely, the lower Aptian, the uppermost Aptian to lowermost Albian, the upper Albian and in the upper Cenomanian, close to the Cenomanian/Turonian boundary (e.g., Jenkyns, 1980; Schlanger et al., 1987; Sliter, 1989; Arthur et al., 1990; Bralower et al., 1993) (Fig. F9). Schlanger and Jenkyns (1976) hypothesized that these OAEs resulted from the vertical expansion of oxygen minimum zones linked to transgressive sea-level pulses and the reduced oxygenation of bottom waters. Others have theorized that oxygen depletion and the deposition of Corg-rich sediments instead was the consequence of paleoceanographic changes such as salinity stratification (Ryan and Cita, 1977; Thierstein and Berger, 1978) and increased flux of Corg from surface productivity or terrestrial sources (e.g., Dean and Gardner, 1982; Parrish and Curtis, 1982; Pedersen and Calvert, 1991).
The deposition of mid-Cretaceous Corg-rich sediments coincided with a worldwide pulse in ocean crustal production (Fig. F10) (Larson, 1991a; Tarduno et al., 1991; Arthur et al., 1991; Erba and Larson, 1991). The release of mantle CO2 from this enormous volcanic episode may have directly caused mid-Cretaceous "greenhouse" warming. The increased preservation and production of organic carbon may have resulted from this warming (e.g., Arthur et al., 1985) combined with increases in nutrients, while sea level rose as the result of the creation of an anomalously young, and therefore shallow ocean floor (Hays and Pitman, 1973; Schlanger et al., 1981).
Regardless of their origin, the burial of Corg-rich sediments enriched in 12C led to significant positive 13C excursions. These have been documented for the Cenomanian/Turonian boundary (Scholle and Arthur, 1980), the early Aptian, and the late Aptian-early Albian (e.g., Weissert, 1989; Bralower et al., 1999). Short-lived negative 13C excursions at the onset of the events may be related to input of mantle CO2 during volcanic events (e.g., Bralower et al., 1994) or to the dissociation of methane hydrates (Jahren and Arens, 1998; Opdyke et al., 1999; Jahren et al., 2001). OAEs are known to be times of rapid turnover among marine biotas as a result of complex changes in habitats (Coccioni et al., 1992; Erba, 1994; Premoli Silva and Sliter, 1999; Premoli Silva et al., 1999; Leckie et al., in press).
Complicating the development of paleoceanographic models are apparent differences in the stratigraphic extent and paleobathymetry of Corg-rich deposits from the Pacific compared to the Atlantic and Tethys Oceans. In the Atlantic and Tethys, Corg-rich deposits occur mostly in basinal settings characterized by major inputs of terrestrial Corg by turbidity currents that led to vertically expanded, long-lived episodes of deepwater anoxia (e.g., Arthur and Premoli Silva, 1982; Arthur et al., 1984; Stein et al., 1986). Terrestrial Corg-rich deposits in the Atlantic and Tethys occur in intervals besides the OAEs (e.g., Bralower et al., 1993). The record of carbonaceous strata in the Pacific is concentrated in the OAEs, dominated by marine Corg, and almost exclusively restricted to paleobathymetric highs (e.g., Dean et al., 1981; Thiede et al., 1982). However, our understanding of the Pacific record is based on scattered occurrences of carbonaceous strata from Shatsky, Hess, and Magellan Rises, the Mid-Pacific Mountains, the Manihiki Plateau, the Mariana Basin, and the accreted oceanic limestone from the Franciscan Complex along the western margin of North America (Sliter, 1984).
Recovery of mid-Cretaceous Corg-rich deposits at relatively shallow burial depth from Shatsky Rise will help determine (1) how sedimentation (i.e., lithology, amount and type of Corg) differs between OAE intervals and non-OAE intervals; (2) the biotic, sedimentologic, and geochemical similarities and differences between the different OAE episodes; (3) whether an oxygen minimum zone model is applicable for OAEs on Shatsky Rise; (4) whether there are any differences between the recovery of Corg-rich sediments on the Northern and Southern Highs that might indicate that the intensity of upwelling differed as a function of latitude; (5) the effect of the worldwide, mid-Cretaceous volcanic pulse on the deposition of Corg and the timing of OAEs; (6) if volcanism was a direct cause of "greenhouse" climate conditions and whether volcanism or methane dissociation is a more likely trigger; and (7) how microplankton and microbenthos responded to the physical, chemical, and biological oceanographic changes associated with OAEs.
The Early and mid-Cretaceous were critical times in the evolution of planktonic foraminifers and calcareous nannoplankton (e.g., Roth, 1987; Leckie, 1989; Premoli Silva and Sliter, 1999; Leckie et al., in press). Nannoplankton underwent dramatic radiations close to the Jurassic/Cretaceous and Barremian/Aptian boundaries (e.g., Bralower et al., 1989, 1994). Both of these events have been documented in the Atlantic and Tethys, but not yet from the Pacific. Pacific sites recording these diversification events would help provide an understanding of their causes.
Planktonic foraminifers appear to have evolved in the Bajocian (Middle Jurassic), but their occurrence is sporadic below the Lower Cretaceous. The diversification of this group was, until recently, thought to have occurred in the early Aptian. Coccioni and Premoli Silva (1994), however, found the evolutionary appearance of a number of taxa far below their previous ranges in the lower Valanginian of the Rio Argos section of Spain. Documentation of this diversification event in other locations and oceanographic settings will help our understanding of its causes.
Shatsky Rise drilling will help us answer the following questions: (1) How did the evolution of nannoplankton correlate to changes in ocean thermal structure and circulation? and (2) Is there evidence for diversification of planktonic foraminifers in the early Valanginian as in Spain, and if so, did this event correlate with any obvious changes in circulation or climate?
A major change in stable carbon isotope ratios of marine carbonates and organic matter has been observed in the Valanginian (e.g., Lini et al., 1992). The event appears to correlate with a major burial event of Corg, an increase in atmospheric CO2, and global warming, perhaps the earliest indications of the Cretaceous "greenhouse" climate (Lini et al., 1992). Increased crustal production rates at this time (e.g., Larson, 1991b) suggest that the event may have a volcanic origin. Warming in the Valanginian is at odds with the evidence of Stoll and Schrag (1996) and others for glaciation in this part of the Cretaceous. Shatsky drilling will help us address how the Valanginian carbon isotope record correlates to indicators of climate change and volcanism and whether there is evidence for warming or cooling in this time interval.
The Early and mid-Cretaceous were characterized by major changes in the level of the CCD (e.g., Thierstein, 1979; Arthur and Dean, 1986). These changes likely resulted from changes in fertility, sea level, ocean floor hypsometry, and ocean circulational patterns. One of the most dramatic events occurred in the early Aptian, at around the same time as the massive Pacific volcanic event, suggesting that volcanism played a direct role, perhaps through increased pCO2. The few data that exist for the Pacific suggest a different CCD history from the Atlantic (Thierstein, 1979), and more data will help resolve the history of the Pacific CCD. Shatsky Rise drilling will help us address the following questions: (1) What was the gradient of carbonate dissolution in the mid-Cretaceous Pacific Ocean? (2) What was the history of variation in the lysocline and CCD in the Early and mid-Cretaceous? and (3) Was a major early Aptian CCD shoaling episode observed for the Atlantic Ocean basins characteristic of the global ocean, or were the oceans out of phase as the result of the pattern of deepwater aging?
LIPs such as the Ontong Java Plateau and Shatsky Rise were constructed during voluminous magmatic events that took place over geologically brief (<1 m.y.) time intervals (e.g., Duncan and Richards, 1991; Tarduno et al., 1991; Coffin and Eldholm, 1994). These events are thought to be associated with massive thermal anomalies in the mantle known as "superplumes" (Larson, 1991b). A likely possibility is that the voluminous phase of superplume activity was associated with the ascent of a plume "head" and that activity declined as the head was depleted of magma and the less voluminous "tail" became the source. One of the major questions concerning the origin of LIPs such as Shatsky Rise is whether they formed in a midplate setting or at a divergent boundary, possibly a triple junction, at times of changing plate geometry (e.g., Sager et al., 1988).
Trace element geochemistry of most samples from Shatsky Rise are close to mid-ocean-ridge basalt (Tatsumi et al., 1998), indicating that they were generated at a divergent boundary, but a few samples have an affinity closer to Polynesian alkalic basalts, suggesting a midplate origin. The latter result is not unexpected, since Shatsky Rise is thought to have formed in the South Pacific (McNutt and Fischer, 1987) near crust with the distinctive Polynesian chemistry. Additional basement drilling at Shatsky Rise will help us address whether it had some kind of a hybrid origin or whether there are other explanations for the few anomalous samples.
Although volcanic basement crops out at several localities on Shatsky Rise (Sliter et al., 1990), basement samples obtained are from dredges, and one pebble in a core catcher sample from DSDP Site 50, and all of these are heavily altered. Ozima et al. (1970) dated volcanic rocks dredged from Shatsky Rise as Tertiary in age. Either these rocks were derived from late stage volcanism identified in seismic data from the rise, or more likely, their pervasive alteration precluded reliable age determination using the K-Ar technique. Maximum estimates for the age of basement on Shatsky Rise can be obtained by adjacent magnetic anomalies (Nakanishi et al., 1989). These ages range from 148 Ma (Tithonian polarity Zone CM21) at the Southern High to 136 Ma (Berriasian-Valanginian polarity Zone CM14) at the Northern High based on the timescale of Gradstein et al. (1994). Fresh basement samples will provide valuable age information.
