Leg 198 was designed to obtain a depth transect of the Cretaceous through Paleogene Pacific Ocean to advance our understanding of the behavior of Earth's climate during "greenhouse" intervals. The Cretaceous and Paleogene are an extended interval of global warmth, on which are superimposed some of the most abrupt and transient climatic changes in the Phanerozoic, including the late Paleocene thermal maximum (e.g., Kennett and Stott, 1991; Zachos et al., 1993), the MME (e.g., MacLeod and Huber, 1996), and the early Aptian OAE1a (e.g., Bralower et al., 1994; Erba et al., 1999). These transient events, less than a million years in duration, are associated with abrupt changes in sea-surface temperatures (SSTs) (e.g., Zachos et al., 1994), switches in the mode of deep sea circulation (e.g., Frank and Arthur, 1999), drastic changes in primary production in surface waters and in marine environments, profound turnover in marine communities (e.g., Erba, 1994; Kelly et al., 1996; Thomas, 1998), and sharp changes in geochemical cycling and burial of organic and inorganic materials (e.g., Arthur et al., 1985; Dickens et al., 1995). A surge in interest in transient climate events has accompanied the recognition that they may have involved rates of input of greenhouse gases similar to modern emissions (e.g., Dickens et al., 1997; Norris and Röhl, 1999).
The last couple of decades of ocean drilling have involved recovery of a number of high-quality records of the mid-Cretaceous and Paleogene interval of global warmth. Stable isotopic and micropaleontological investigations of these sections allowed us to compile detailed long-term records of sea-surface and deep-water temperatures and to identify times of peak warmth, as well as transitional intervals into and out of this "greenhouse" interval (e.g., Huber et al., 1995; Jenkyns et al., 1995; Clarke and Jenkyns, 1999; Norris and Wilson, 1998; Erbacher et al., 2001; Wilson and Norris, 2001). Combination of data from low- and high-latitude sites can help us determine thermal gradients; however, we can still only speculate about the nature of deep sea circulation. Moreover, we still do not have a complete understanding of how global warmth was maintained for long periods of time with apparently low thermal gradients (e.g., Sloan and Thomas, 1998; Huber and Sloan, 2000), nor do we completely understand how these thermal gradients affected deep sea circulation patterns. We have also yet to determine how the ocean maintains apparently high or low fertility states for extended intervals and how this affected the balance of key nutrients. Finally, we do not fully understand the detailed mechanisms that transform marine communities during times of abrupt environmental change.
One of the largest obstacles facing our understanding of Cretaceous and Paleogene climate and ocean circulation is that many good stratigraphic sections have been buried at depths where diagenetic alteration has obscured stable isotope and other climate proxies. In other sequences, spot-coring, coring gaps, and drilling disturbance hinder detailed paleoceanographic studies. The result is that site coverage is uneven and almost nonexistent in some areas. This is especially the case for the Pacific Ocean. The areal extent and importance of the Pacific in global circulation, however, makes it a critical target for drilling of warm climatic intervals.
One of the most promising locations in the Pacific for recovering Cretaceous and Paleogene sediments at relatively shallow burial depths is Shatsky Rise, a medium-sized LIP, in the west-central Pacific (Fig. F1). Shatsky Rise has been the target of three DSDP expeditions—Legs 6, 32, and 86—and ODP Leg 132. The highest quality record was obtained during Leg 86 at Site 577, which was limited to the Paleogene and uppermost Maastrichtian. Some sites in the older legs were spot-cored, and chert has lowered recovery in others, especially in the Cretaceous. Rotary coring of the unindurated Cenozoic section has led to considerable disturbance. Yet even with an extremely patchy record, SST estimates obtained from isotopic analyses of Shatsky Rise sediments have provided scarce but important data for our understanding of Cretaceous and Paleogene climates. The Leg 198 depth transect was designed to provide additional data from the tropical Pacific as well as a representative open-ocean signal.
The majority of sites used in current paleoceanographic investigations were situated at relatively shallow paleodepths, mostly less than 2000 m. The selected Shatsky sites provide a depth transect spanning 1500 m of the water column, providing us the opportunity to sample true intermediate and deep waters. Even though the paleodepth estimates for the Leg 198 sites are somewhat uncertain, their broad depth range ensures that the deepest sites will lie below previously drilled sites. Existing recovery from Shatsky demonstrates that pristine isotope-grade material will be recovered for most of the entire interval of interest due to its relatively shallow burial depth. This will allow the following specific objectives to be addressed:
Finally, drilling of basement on Shatsky Rise will provide samples to determine the age and composition of basement under the rise. The compositional data from Shatsky Rise basement and other LIPs will be compared in light of the different models for their formation.
Shatsky Rise, a broad elevation in the west-central Pacific, is the oldest existing oceanic plateau. The rise is 1650 km long, 450 km wide, and consists of three prominent highs (Southern, Central, and Northern) arranged in a southwest-northeast trend (Fig. F2). The Southern High, also known as Shatsky Plateau, is the largest high with a length of about 700 km and a width of about 300 km. Central and Northern Highs are less than half the size of the Southern High. The highs have tops at 2.5–3.5 km and gentle flanks that rise from the surrounding seafloor at 5.5–6 km. This topography is consistent with production by effusive flood basalt–style volcanism. The highs are volcanic edifices that are surrounded by normal oceanic lithosphere and a group of about 80 seamounts (Sager et al., 1999). Although there has been no basement drilling of Shatsky Rise, altered basalts have been dredged (Sager et al., 1999).
Southern High is elongated southwest to northeast with a subcircular summit that rises to ~2400 m (Fig. F2). The sediment pile on the high is up to ~1200 m thick. Jurassic/Cretaceous rudists and corals have been dredged from the south edge of the Southern High summit, suggesting that the Southern High was emergent early in its history (Sager et al., 1999).
Central High is subcircular with a broad summit rising to depths of ~3000 m Nakanishi et al., 1999). The sedimentary section is up to 1 km in thickness. Underneath the sediments are a series of basement ridges and faults. The Central High flanks appear to follow lineation and fracture patterns, and thus were likely shaped by ridge tectonics (Sager et al., 1999). The smallest of the highs, Northern High, has a domed summit rising to 3100 m. The sediment package on this high reaches about 1 km in thickness (Sager et al., 1999). The shape and orientation of the flanks also suggest that they were formed by ridge tectonics (Nakanishi et al., 1999; Sager et al., 1999).
Shatsky Rise intersects the southwest-northeast trending Japanese magnetic lineations and the northwest-southeast trending Hawaiian lineations (Nakanishi et al., 1989). Thus the regional magnetic anomalies of the abyssal Pacific seafloor surrounding Shatsky Rise exhibit a nearly orthogonal pattern (Sager et al., 1988; Nakanishi et al., 1992) with the presumed intersection of the anomalies near the crest of the rise. The observation of lineations continuing through Shatsky Rise indicates that it formed on oceanic lithosphere (Nakanishi et al., 1999).
The rise is bracketed by magnetic polarity Zones M21 and M10N (Fig. F3), with a progression of ages and decreasing volume of highs from southwest to northeast (Southern to Northern High) (Sager and Han, 1993). Because the rise lies at the triple junction of the Hawaiian and Japanese magnetic lineation sequences, its formation is thought to be related to the evolution of the junction between them (e.g., Larson and Chase, 1972; Hilde et al., 1976). Magnetic lineations cross only the lower part of the three highs, suggesting that they are volcanic edifices separated by normal lithosphere (Nakanishi et al., 1999). Nakanishi et al. (1989) proposed that Shatsky Rise formed by a magmatic pulse at a hotspot triple-junction intersection beginning in polarity Chron CM21 in the Tithonian (Late Jurassic) and ending at ~ polarity Chron CM12 in the Valanginian (Early Cretaceous). Age gaps between the highs suggest that plateau volcanism was episodic. Sager and Han (1993) noted uniform reversed magnetization on the Southern Rise and proposed that it formed rapidly during one reversed polarity chron, although the calculated rate of formation was somewhat slower than Ontong Java Plateau (Tarduno et al., 1991). The rise was likely formed by eruption of a large igneous pile during initial plume head activity (Sager and Han, 1993; Sager et al., 1999).
A number of lineations form magnetic bights in the middle of the rise. These bights indicate past locations of triple junctions (Nakanishi et al., 1999; Sager et al., 1999). Bights older than M22 occur south of Shatsky Rise. Bights between M19 and M14 are found between the Southern and Northern Highs. Gaps between magnetic lineation bights indicate large triple-junction jumps (Nakanishi et al., 1999). It is possible that the triple-junction movement was caused by ridge jumps as the triple junction moved to stay over the plume, which drifted northeast relative to the Pacific plates as the plates moved over the mantle. The formation of Shatsky Rise occurred about the same time as an 800 km eastward jump of the Pacific-Izanagi-Farallon Triple Junction and reorganization of Pacific-Izanagi Ridge between Chrons CM21 and CM19. The ridge jump was possibly caused by the eruption of Shatsky Rise. Other ridge jumps occurred around the time of the formation of the Central and Northern Highs (Sager et al., 1999).
Although paleoreconstructions have a great deal of uncertainty, Shatsky Rise appears to have formed in equatorial latitudes in the Southern Hemisphere and gradually drifted northward during the last 90 m.y. (Larson et al., 1992) (Fig. F4).
All previous DSDP and ODP drill sites are located on the Southern High (Fig. F3). The sedimentation history of the Central and Northern Highs was not known prior to Leg 198. Pelagic sedimentation on the Southern Rise began in the latest Jurassic or earliest Cretaceous and shows a moderately stratified section overlying acoustic basement. The section is up to 1.2 km thick in places (Ewing et al., 1966; Sliter and Brown, 1993). Sedimentation appears to have been interrupted by episodic, regional, or local erosional events identified by previous drilling (Sliter and Brown, 1993). Some of the unconformities show up as major reflectors observed in seismic records. Quaternary channeling of the sediment pile is evident on seismic records, and in some places Lower Cretaceous sediments crop out at the seafloor (Sliter et al., 1990), for example near Site 306 (Site 1214). Seismic profiles show abrupt changes in stratigraphy over short distances.
Divergence of deep reflectors suggests that there was more rapid basement subsidence in the Early Cretaceous prior to the deposition of Upper Cretaceous and Cenozoic sediments. However, the subsidence history of Shatsky Rise is not well known. Thierstein (1979) backtracked Sites 305 and 306 (currently at 2903 and 3416 m, respectively) to close to 1000 m paleodepth in the Early Cretaceous sediments. Recently, rudists, corals and echinoid spines were dredged from a ridge on the Southern High at 3000 m (Sager et al., 1999), suggesting paleodepths close to sea level. Detailed studies of benthic foraminiferal assemblages are required to more accurately constrain the subsidence history of Shatsky Rise and compare it to other features in the Pacific with a clearer subsidence history.
A total of eight previous DSDP and ODP drill sites lie on the Southern High (Fig. F3). At Site 47 atop the southwest flank of the Southern High, the highly discontinuous, ~130-m Cenozoic and Maastrichtian section that was recovered consists of ooze and chalk. Nearby, at Site 48, the ~85-m section contains Miocene ooze resting directly on Maastrichtian chalk (Fischer, Heezen, et al., 1971). At Site 49, on the lower southwestern flank, a thin (~5 m) Pleistocene zeolitic clay unit sits directly on ~15 m of Tithonian–Berriasian cherty chalk. The nearby section at Site 50 includes ~35 m of Pleistocene zeolitic clay and siliceous ooze overlying ~9 m of Tithonian–Berriasian chalk. At Site 305, on the southern flank of Southern High, 640 m of section was cored. The section includes Campanian to Holocene ooze and chalk with minor, disseminated chert overlying Santonian to Coniacian and Cenomanian to Barremian chalk with regular chert nodules (Larson, Moberly, et al., 1975; Sliter, 1992). Nearby, drilling at Site 306 recovered Holocene ooze directly overlying chalk of Albian age. Alternating chalk and chert extend down to the Berriasian, where the hole was terminated at 475 meters below seafloor (mbsf) (Fig. F5). Scraps of Corg-rich sediments of early Aptian age were recovered at Sites 305 and 306 (Larson, Moberly, et al., 1975). Sliter (1989) refined the biostratigraphy of the mid-Cretaceous of Site 305 and placed the organic-rich units in the Globigerinelloides blowi planktonic foraminiferal zone, correlating them with OAE1a, the lowermost mid-Cretaceous OAE (Bralower et al., 1993). Additional fragments of carbonaceous shale from Core 37 at Site 305 are early Cenomanian to early Turonian in age and apparently underlie a major unconformity that extends to the Coniacian.
Drilling at Site 577, which is located near Site 47, recovered ~120 m of nannofossil ooze (Heath et al., 1985). The section includes a thin upper Miocene to Holocene interval overlying a relatively expanded upper Maastrichtian–Paleocene interval. The K/T boundary at Site 577 is nearly complete biostratigraphically and contains an Ir anomaly (Wright et al., 1985). The boundary interval has been the subject of a host of paleoceanographic investigations (e.g., Gerstel et al., 1986; Zachos and Arthur, 1986; Zachos et al., 1989). The LPTM lies between cores at Site 577 (Pak and Miller, 1992). Site 810, the most recently cored Shatsky site that was drilled during Leg 132 primarily as an engineering test, recovered a highly discontinuous Cenozoic and upper Maastrichtian section composed of ooze, chalk, and chert (Storms, Natland, et al., 1991; Premoli Silva et al., 1993).
Two regional reflectors or sets of reflectors divide the sedimentary column of the Southern High. Sliter and Brown (1993) defined R1 as a prominent unconformity near the Cenomanian/Turonian boundary and R2 as a prominent chert horizon that lies close to the Barremian/Aptian boundary. Less prominent reflectors occur near the Paleogene/Neogene and K/T boundaries. Sliter and Brown (1993) also divided the Southern High stratigraphic section into five main stratigraphic units, a scheme that is also applied here: Units 1 and 2 are Neogene and Paleogene in age, respectively; Unit 3 is Turonian–Maastrichtian (between R0 and R1); Unit 4 is Aptian–Cenomanian (between R1 and R2); and Unit 5 is Berriasian–Barremian (below R2). The sedimentary sections on the Central and Northern Highs can be very tentatively correlated with the Southern High section based on the major reflectors. However, the absence of thick sedimentary sections between the highs lessens the credibility of these correlations (A. Klaus, W. Sager, and L. Khankishieva, pers. comm., 1997). Thus, the detailed sedimentary history of the Central and Northern Highs remained largely unknown prior to Leg 198.
The sedimentary history of Shatsky Rise appears to have been greatly affected by crossing the equatorial high-productivity zone in the mid- to Late Cretaceous (Fig. F4). This crossing led to the accumulation of substantial amounts of siliceous material that was concentrated in chert horizons during burial.
Shatsky Rise lies in the path of surface-water currents transporting two very different water masses. The first, the Kuroshio Extension, is the arm of the Kuroshio Current extending eastward into the Pacific Ocean off the coast of Japan. This current is the Pacific equivalent of the Gulf Stream after it leaves the coastline of North America (Kawai, 1970). The second current, the Oyashio, transports water from the subarctic water mass. These currents converge near the Northern High. However, depending on the exact location of the currents, which changes significantly over time, temperature and current velocity vary substantially over the rise. Average salinity for these surface waters ranges between 34 and 35 g/kg.
The Kuroshio and Oyashio Currents impose structure on underlying intermediate waters, marking a strong vertical gradient in temperature and salinity. At 1000 m depth, temperatures average 3°C. These waters also have salinities near 34 g/kg. The intermediate waters flow toward the east as part of a clockwise gyre in the North Pacific.