Previous   |   Next

doi:10.2204/iodp.proc.320321.101.2010

Site summaries

Site U1331

Three holes were cored at Site U1331 (12°04.088′N, 142°09.708′W; 5116 mbsl), which is the northwesternmost site drilled during the PEAT program (Fig. F53; see Table T1 in the "Site U1331" chapter). In Hole U1331A, seafloor basalt is overlain by 187.02 m of pelagic sediment, containing radiolarian and nannofossil ooze with varying amounts of clay. The oldest sediment is of early Eocene age. For detailed coring activities, see "Operations" in the "Site U1331" chapter.

The sediment column at Site U1331 has a strong resemblance to that of Site 1220 (Fig. F54) (Lyle, Wilson, Janecek, et al., 2002), but with noteworthy sharp erosive contacts concentrated within the upper two-thirds of the section (Fig. F55). A total of 7 m of Pleistocene–Pliocene clay overlies lower Oligocene to lower Eocene alternations of nannofossil ooze, radiolarian nannofossil ooze, and radiolarian oozes of varying clay and calcium carbonate content (see "Lithostratigraphy" in the "Site U1331" chapter), with a sharp lithologic change at the Eocene–Oligocene transition (~26 m core depth below seafloor [CSF]). The lowermost target interval in Hole U1331C is characterized by a ~20 m thick radiolarian ooze and porcellanite layer from 157 to 177 m CSF overlying a dominantly clayey radiolarian ooze and zeolitic clay with hydrothermal red staining from 177 to 187.5 m CSF deposited on top of a thin (187.6–188.5 m CSF in Hole U1331C) layer of calcareous ooze and zeolitic clay above the basalt. Some fine-grained basaltic fragments show fresh glassy chilled margins.

Carbonate content approaches 80 wt% in the Oligocene nannofossil oozes in the upper part of the site and cycles between 0 and 40 wt% in the middle Eocene section (see "Geochemistry" in the "Site U1331" chapter). A concentration of sharp erosive contacts occurs in the interval between 80 and 120 m CSF, with calcareous material dominating the basal portion of these contacts and then fining upward in grain size into radiolarian oozes. Rarely, the sediment above a sharp contact contains well-rounded clasts up to 1 cm in diameter (interval 320-U1331B-10H-6, 117–130 cm). The lithologic and stratigraphic characteristics of these sediments have been interpreted as gravity-driven deposits, possibly from nearby seamounts ~10 km to the south (Fig. F53B). Between ~177 and 187.2 m CSF, Cores 320-U1331A-22X, 320-U1331C-16H, and 17H achieved our site objective of recovering carbonate-bearing material from the time interval just following 52 Ma.

All major microfossil groups were found in sediments from Site U1331, and they provide a consistent and coherent biostratigraphic succession from basement to below the surficial clay layer. Nannofossils are common in the Oligocene and lower Eocene but sporadic in sediments from the upper Eocene because of dissolution. Middle Eocene sediments commonly contain calcareous nannofossils punctuated by several barren intervals, notably below Zone NP21 (radiolarian Zone RP19 equivalent), below Zone NP17, and between Zones NP15 and NP13 (radiolarian Zones RP12–RP8 equivalent). Radiolarians are common to abundant throughout the section. Radiolarian and nannofossil datums and their age determinations agree and range from nannofossil Zone NP12 in the basal sediment section (~51–53 m.y. before present) to Zones NP23/NP24 and radiolarian Zones RP8 just above basement through RP21 (late Oligocene, older than 25 Ma) in the uppermost section, below the Pleistocene clays. Both radiolarian and nannofossil assemblages contain reworked older components (deeper than ~50 m CSF) but within a coherent and ordered stratigraphy. Planktonic foraminifers are generally absent, except for sporadic occurrences often associated with sediment just above sharp lithologic contacts and also in the basal carbonate section (Zones E4/E5). Benthic foraminifers are generally rare and indicate lower bathyal to upper abyssal paleodepths. They are frequently found in the graded coarse sediment above the base of sharp contacts but indicate there is no apparent difference in the depth habitat between benthic foraminifers from just above sharp contacts and other parts of the section.

Apparent sedimentation rates, as implied by the biostratigraphic age determinations and aided by magnetostratigraphic polarity interpretations, vary throughout the section. The radiolarian-rich section between ~80 m CSF and basement was deposited at an average rate of 10 m/m.y., whereas the late middle Eocene to Oligocene section was deposited at a rate of ~4 m/m.y., with an apparent inflection between 60–80 m CSF. Porcellanite is located in an interval that spans a time of ~2–3 m.y. The presence of all major fossil groups as well as a detailed magnetostratigraphy will allow us to achieve one of the main PEAT objectives to arrive at an integrated Cenozoic stratigraphy and age calibration (e.g., Pälike et al., 2006b).

A full physical property program was run on cores from all three holes, including Whole-Round Multisensor Logger (WRMSL) measurements of magnetic susceptibility, bulk density, P-wave velocity, and noncontact resistivity, along with natural gamma radiation (NGR), followed by discrete measurements of color reflectance, index moisture and density properties, sound velocities, and thermal conductivity. Bulk density measurements increase markedly in the carbonate-rich Oligocene section. Magnetic susceptibility measurements are variable throughout the section, allowing detailed correlation between different holes and picking of sharp contacts and clay layers by increased susceptibilities. NGR measurements are elevated by an order of magnitude in the surficial clay layer and reach 130 cps at the seafloor, dropping to <5 cps deeper than 30 m CSF. Porosity values are generally high in radiolarian-rich sediments (80%) and decrease within the Oligocene carbonate section. Carbonate content is positively correlated with thermal conductivity. Discrete physical property measurements will prove useful to calibrate WRMSL velocity and density estimates and generally agree with WRMSL estimates, once appropriate correction factors are included for the core liner. Discrete velocities are significantly higher (50–100 m/s) than track measurements in the direction perpendicular to the split plane of the core section (x-axis), which is likely an artifact.

Using whole-round magnetic susceptibility measurements, Holes U1331A–U1331C can be spliced to form a continuous section to at least 140 m CSF or 150 m core composite depth (CCSF-A) (see "Core composite depth scale" in the "Methods" chapter), with no apparent gaps. Core expansion is ~15%. It is possible that Hole U1331C cores can provide an additional spliced section to the top of the porcellanite interval at ~157 m CSF. Below 149 m CCSF-A, it was only possible to tentatively correlate features in the track data down to Core 320-U1331A-17X for a total composite section to ~172 m CCSF-A.

A full range of paleomagnetic analyses was conducted on cores and samples from Site U1331. Our aims are to determine the magnetostratigraphy and study geomagnetic field behavior, environmental magnetism, and Pacific plate paleogeography. Shipboard analyses conducted suggest that a useful magnetic signal is preserved in almost all APC-cored intervals. Preliminary comparison of biostratigraphic data and changes in magnetic paleodeclinations suggest the recovery of Oligocene magnetochrons to the base of the middle Eocene (Chron C21n; ~47 Ma). Paleomagnetic directions from discrete samples agree well with those from split-core results.

A standard shipboard suite of geochemical analyses of pore water and organic and inorganic sediment properties was conducted on samples from Site U1331, including a pilot study of high-resolution "Rhizon" pore water sampling, which does not require the cutting of whole rounds for squeezing. Carbonate coulometry yielded carbonate concentrations of ~80 wt% in the Oligocene nannofossil ooze and sporadic horizons with up to 40 wt% CaCO₃ in the middle Eocene radiolarian-rich oozes. Preliminary calcium carbonate determinations from the white, hydrothermally stained sediments just above basement (whole-round Sample 320-U1331C-17H-4, 83–84 cm) yielded low values of only 2–3 wt% CaCO₃. Alkalinity values range between 2.5 and 3 mM throughout the section. Additional ephemeral samples were taken for shore-based microbiology and permeability studies.

Wireline logging provided valuable information to constrain the interval of porcellanite formation within the borehole, and further analysis will aid in interpretations of carbonate content and lithologies. Integration with the seismic data will allow further improvements with the regional seismic interpretations. Data from Site U1331 indicate that the top of seismic Horizon P2 (Lyle et al., 2002) correlates with the top of the chert section. Downhole temperature measurements with the APCT-3 tool, when combined with the thermal conductivity values obtained from the cores, indicate that Site U1331 had a thermal gradient of 13.4°C/km and a heat flow of 10.3 mW/m². This is within the range of lower values in the global heat flow data set for the eastern Pacific but significantly lower than values obtained for Sites 1218 and 1219.

Highlights

Carbonate ooze in basal section

At Site U1331 we recovered a 1.2 m thick interval (lithologic Unit IV) of calcareous ooze with concretions and reddish color streaks, achieving one of the objectives for this site. The nannofossil ooze recovered in Sample 320-U1331A-22X-CC contains a moderately to poorly preserved assemblage of early Eocene planktonic foraminifers (planktonic foraminifer Zone E5). This moderately preserved assemblage was also observed in the basal section of Hole U1331C.

Stratigraphic integration

One of the primary objectives of the PEAT science program is the integration of different stratigraphic methodologies and tools. Site U1331 contains almost all major fossil groups (nannofossils, radiolarians, foraminifers, and diatoms), as well as an excellent magnetostratigraphy. The possibility of a cycle-by-cycle match between Sites U1331 and 1220 has been demonstrated using magnetic susceptibility and bulk density data, providing additional stratigraphic tie points and verification of the completeness of the stratigraphic section on a regional scale. Thus, Site U1331 will help us to achieve an integrated stratigraphy for the Cenozoic Pacific Ocean.

Eocene–Oligocene transition and depth transect

Site U1331 forms the oldest and deepest component of the PEAT depth transect component, which will allow the study of critical intervals (such as the Eocene–Oligocene transition; see Coxall et al., 2005) and variations of the equatorial CCD. Site U1331 is estimated to have been ~4.2 km deep during the Eocene–Oligocene transition, ~800 m shallower than today. Sediments rapidly change from radiolarian ooze below the Eocene–Oligocene transition to nannofossil oozes above and will provide a tie point for calcium carbonate burial at ~5° paleolatitude.

Variations in the CCD

Site U1331 will provide important constraints for variations and depth of the CCD from the early Eocene to the late Oligocene. This site shows increased carbonate content and much increased mass accumulation rates approximately from the middle of Chron C18r to the base of Chron C19r during the middle Eocene and can be correlated to an interval of enhanced carbonate burial that was previously documented by Lyle et al. (2005) in Leg 199 cores.

Age transect of seafloor basalt

At Site U1331 we recovered what appear to be fresh fragments of seafloor basalt, aged between 52 and 53 Ma as estimated from biostratigraphic results. This material will, when combined with other PEAT basalt samples, prove to be important for the study of, for example, seawater alteration of basalt, as well as paleomagnetic studies.

Site U1332

Three holes were cored at Site U1332 (11°54.722′N, 141°02.743′W; 4924 mbsl) (Fig. F56; see Table T1 in the "Site U1332" chapter), which is the second northwesternmost site drilled during the PEAT program. At Site U1332, seafloor basalt is overlain by ~150 m of pelagic sediment, containing radiolarian and nannofossil ooze with varying amounts of clay and zeolitic clay. The oldest sediment is of earliest middle Eocene age. Hole U1332A provided high-quality and high-recovery APC-cored sediments from the mudline to 125.9 m CSF (Core 320-U1332A-14H), where we encountered porcellanite and chert and switched to the XCB cutting shoe. XCB coring advanced to 152.4 m drilling depth below seafloor (DSF) through a ~10 m thick porcellanite-rich interval with reduced recovery. In the basal section, we recovered a short, ~3.8 m long interval of barren very dense and stiff clay above basalt, ~10 m shallower than predicted from the seismic profile, in Core 320-U1332A-18X. Basement was reached at 152.4 m CSF. For detailed coring activities, see "Operations" in the "Site U1332" chapter.

The uppermost 17.7 m consists of upper Miocene to Pleistocene–Pliocene clay, with varying amounts of radiolarians and zeolite minerals, overlying ~130 m of Oligocene to middle Eocene nannofossil and radiolarian ooze with porcellanite deep in the section. A thin ~3 m thick unit of middle Eocene zeolite clay bearing small porcellanite and chert nodules was recovered at the base of the sedimentary sequence, above basaltic basement. The sedimentary sequence at Site U1332 was divided into five major lithologies (Fig. F57).

The upper stratigraphy at Site U1332 has a strong resemblance to that of Site U1331 but without the sharp erosive contacts described at Site U1331. Several meters of white to beige-colored Pleistocene–Pliocene clay (lithologic Unit I) overlie lower Miocene to lowermost Oligocene nannofossil ooze (Units II and III). There is a sharp lithologic change at the Eocene–Oligocene transition to alternating radiolarian ooze with nannofossils and nannofossil ooze (Subunit IVa). The lithology gradationally changes downhole into a dominance of radiolarian nannofossil ooze and nannofossil radiolarian ooze (Subunit IVb) and then into an interval of alternating radiolarian ooze, radiolarian nannofossil ooze, and nannofossil radiolarian ooze with porcellanite layers (Subunit IVc). Lithologic Unit V is composed of very dark grayish brown to black clay, very dark grayish brown to black zeolite clay, and chert. The sediments directly above basaltic basement are partially lithified. Basalt is designated as lithologic Unit VI, at ~150 m CSF.

Carbonate content approaches 85 wt% in Unit III within the Oligocene nannofossil oozes and cycles between 0 and 40–60 wt% in the middle Eocene section (Unit IV) (Fig. F58). All major microfossil groups were found in sediments from Site U1332 and provide a consistent, coherent, and high-resolution biostratigraphic succession from basement to the top of Unit II. Calcareous nannofossils are abundant and moderately well preserved in the Oligocene and poor to moderately well preserved in the Miocene and Eocene. Most middle Eocene sediments commonly contain nannofossils; however, there are several barren intervals. Radiolarians are common to abundant throughout most of the section, apart from the lowermost sediment section above basalt, and are well preserved, particularly in the Eocene and in the early Oligocene. Radiolarian and nannofossil datums and zonal determinations agree, ranging from nannofossil Zones NP13/NP14 in the basal dark clay section (~48.4–50.7 Ma) to Zone NN1 and radiolarian Zones RP13 above basement through RN1 (lowermost Miocene, ~22.3 Ma) below the upper Pliocene–Pleistocene clay cover in Core 320-U1332A-3H (Fig. F58). Planktonic foraminifers are generally rare throughout the Oligocene but are absent in the Miocene and Eocene. Benthic foraminifers are present through most of the section but are rare in Miocene and Eocene sediments. They indicate lower bathyal to abyssal paleodepths. Sedimentation rates, as implied by biostratigraphic age determinations, vary throughout the section and are ~5 m/m.y. in the Eocene section and ~2.5 m/m.y. in the Oligocene, with two prominent hiatuses in the Miocene and between the Miocene and younger sediments. The presence of all major fossil groups as well as a detailed and well-resolved magnetostratigraphy will allow us to achieve one of the main PEAT objectives, to arrive at an integrated Cenozoic stratigraphy and age calibration (e.g., Pälike et al., 2006b) for major parts of the Oligocene and Eocene.

Magnetostratigraphic studies as well as high-resolution biostratigraphy and stratigraphic correlation determined that a 4 m interval from the base of Core 320-U1332A-8H was repeated in the top of Core 9H, which comprises Chron C13n and the lowermost Oligocene. This repetition also occurs in Cores 320-U1332B-8H and 9H and within Core 320-U1332C-9H. The lithologic succession from the lower occurrence of Chron C13n downhole as well as from the upper occurrence of Chron C13n uphole both appear complete and continuous; hence Site U1332 achieved the fortuitous feat of recovering the complete Eocene–Oligocene transition four times and the upper part of Chron C13n five times at a triple-cored site. A likely explanation for this is the widespread occurrence of a slumped interval.

A full physical property program was run on cores from all three holes, including WRMSL measurements of magnetic susceptibility, bulk density, P-wave velocity, and noncontact resistivity, along with NGR, followed by discrete measurements of color reflectance, index moisture and density properties, sound velocities, and thermal conductivity. Bulk density measurements show a marked increase in the carbonate-rich Oligocene section, as well as in carbonate-bearing horizons in the Eocene (CAE cycles; Lyle et al., 2005). Magnetic susceptibility is variable throughout the section, allowing a detailed correlation among holes. NGR measurements are elevated by an order of magnitude in the surficial clay layer. Porosity values are generally high in the radiolarian-rich sediments (85%) and decrease in the Oligocene and Eocene carbonate section, which also shows higher thermal conductivity values of ~0.9 to 1.2 W/(m·K), compared with ~0.8 W/(m·K) in the radiolarian oozes and surficial clay.

Stratigraphic correlation allowed us to obtain a complete section to ~125.5 m CSF near the top of the porcellanite interval in Hole U1332A, equivalent to a composite depth of ~140 m CCSF-A. The overall core expansion (growth factor), which is calculated by the ratio between the CCSF-A and CSF (formerly meters composite depth [mcd] and meters below seafloor [mbsf]) depth scales, is ~10%. The tops of APC cores were often affected by ~3 m heave that occurred during operations at Site U1332. Stratigraphic correlation supports the biostratigraphic, paleomagnetic, and sedimentologic description of a repeated sequence, possibly due to slumping, spanning the Eocene–Oligocene transition.

A full range of paleomagnetic analyses was conducted on cores and samples from Site U1332 and resulted in a well-resolved magnetostratigraphy. Shipboard analyses suggest that a useful magnetic signal is preserved in all APC-cored intervals and that it was possible to remove the drilling-induced steep inclination overprint by AF demagnetization. Comparison of biostratigraphic data and changes in magnetic paleodeclinations suggests the recovery of magnetic reversals Chrons C1n/C1r.1r to C2An.3n/C2Ar above a hiatus and then a continuous sequence of magnetic reversals from Chrons C5En/C5Er (18.52 Ma) in the Miocene at ~12.95 m CSF (interval 320-U1332C-2H-4, 95 cm) to C19r/C20n (42.54 Ma) at interval 320-U1332A-14H-5, 80 cm. Magnetostratigraphic interpretation supports the presence of a slump through multiple recovery (five times) of parts of Chron C13n in a triple-cored sequence. Paleomagnetic directions from discrete samples agree well with those from split-core results.

A standard shipboard suite of geochemical analyses of pore water and organic and inorganic sediment properties was conducted on samples from Site U1332. Alkalinity values increase from ~2.2 to 3.4 mM downsection, and Sr²⁺ increases from ~80 to ~110 µM. H₄SiO₄ remains relatively stable between 400 and 600 µM above 90 m depth in the Oligocene nannofossil oozes but increases to 800–1000 µM in the Eocene silica-rich radiolarian oozes. Carbonate coulometry yielded carbonate contents of ~85 wt% in the Oligocene nannofossil ooze and horizons with up to 60 wt% CaCO₃ in the middle Eocene radiolarian-rich oozes. TOC contents were measured both by difference between TC and total IC as well as by using an acidification method. Using the acidification method, TOC values were <0.3 wt% for all measured samples. The top ~5 m shows values of 0.18–0.17 wt% TOC. Between ~40 and 70 m CSF the measurements indicate TOC below the detection limit of 0.03 wt%, and downhole from this, values are generally low. We conducted a high-resolution Rhizon pore water experiment across an alkalinity trough around 40 m CSF, which highlighted comparisons between squeezed and Rhizon-sampled pore waters. Additional ephemeral samples were taken for shore-based microbiology and permeability studies.

Wireline logging provided valuable information to constrain the interval of porcellanite and chert formation within the borehole. Downhole NGR, density, and magnetic susceptibility logs provide important constraints on the poorly recovered lithologies below and between porcellanite-bearing horizons. The logging data document the presence of two thin porcellanite horizons at ~126 and 130 m wireline log depth below seafloor (WSF) and an ~14 m thick interval of increased magnetic susceptibility, reduced conductivity, and enhanced density and photoelectric factor that appears to be the dark and dense clays and zeolitic clays above basement, rather than carbonate. Integration with the seismic data will allow further improvements with the regional seismic interpretations. Data from Site U1332 indicate that the top of seismic Horizon P2 (Lyle et al., 2002) correlates with the top of the porcellanite section, just as it did for Site U1331. No FMS data were collected, as it was not possible to retrieve the "paleo-" triple-combination (triple combo) tool string back into the bottom-hole assembly (BHA). Eight downhole temperature measurements were conducted in Holes U1332B and U1332C with the APCT-3 tool. Three of these yielded good data; the other measurements were impaired by strong, sometimes >3 m heave during operations in Hole U1332B.

Downhole temperature measurements, when combined with the thermal conductivity values obtained from the cores, indicate that Site U1332 has a heat flow of 70.7 mW/m² and a thermal gradient of 75.0°C/km. This is significantly higher than the values obtained for Site U1331 but comparable to values obtained for Sites 1218 and 1219.

Highlights

Shallow early Eocene CCD

Coring at Site U1332 was designed to capture a very short period of time (~2 m.y.) at ~50 Ma during which this site was thought to be located above the very shallow Eocene CCD (~3.3 km) (Lyle, Wilson, Janecek, et al., 2002; Rea and Lyle, 2005) just after the EECO (Zachos et al., 2001a). Unlike Site U1331, at Site U1332 we cored a ~10 m thick section of dense and dark brown clays, zeolite clays, and chert above basement, although relatively common nannofossils were present in the lowermost samples from Hole U1332B. This finding will provide important new constraints on the depth of the CCD at ~48–50 Ma at the paleoequator, indicating that the CCD was shallower than previously thought.

Stratigraphic integration

One of the primary objectives of the PEAT science program is the integration of different stratigraphic methodologies and tools. Site U1332 contains all major fossil groups (nannofossils, radiolarians, foraminifers, and diatoms), as well as an excellent magnetostratigraphy and composite depth correlation, which can be tied to nearby Leg 199 sites (e.g., Site 1220) by way of physical property variations. The possibility of a cycle-by-cycle match between Sites U1332 and 1220 has been demonstrated using magnetic susceptibility and bulk density data, providing additional stratigraphic tie points and a verification of the completeness of the stratigraphic section on a regional scale. Thus, Site U1332 will help us to achieve an integrated stratigraphy for the Cenozoic Pacific Ocean, ranging from the Miocene to the middle Eocene.

Eocene–Oligocene and Oligocene–Miocene transitions and depth transects

Site U1332 forms the second oldest and deepest component of the PEAT depth transect, which will allow the study of critical intervals (such as the Eocene–Oligocene transition; see Coxall et al., 2005) and variations of the equatorial CCD. Site U1332 is estimated to have been ~4 km deep during the Eocene–Oligocene transition, ~1 km shallower than today and 200 m shallower at that time than Site U1331. Sediments rapidly change from radiolarian ooze below the transition into nannofossil oozes above, and unlike Site U1331, Site U1332 also contains carbonate-bearing sediments across the Oligocene–Miocene transition. For the Eocene–Oligocene transition, Site U1332 will provide a tie point for calcium carbonate burial at ~4° to 5° paleolatitude.

Variations in the CCD

Site U1332 has provided important constraints for variations and depth of the CCD from the early Eocene to the late Miocene. This site shows increased carbonate content and much increased mass accumulation rates approaching 200 mg CaCO₃/cm²/k.y. around the middle of Chron C18r to the base of Chron C19r during the middle Eocene, which can be correlated to an interval of enhanced carbonate burial that was previously documented by Lyle et al. (2005) in Leg 199 cores. The early Oligocene high CaCO₃ concentrations decrease significantly in sediments younger than ~27 Ma. By ~22 Ma, in the early Miocene, carbonate was no longer preserved. This is presumably related to Site U1332 sinking below the prevalent CCD and coincides with a CCD shoaling event between ~20 and 15.5 Ma described by Lyle (2003).

Formation of porcellanite and chert

Together with Site U1331, Site U1332 provides important new information on the formation of porcellanite and chert. Coring has shown that the top of the porcellanite-rich interval is mapped by seismic Horizon P2 (Lyle et al., 2002). In lithologic Subunit IVc, layers and pebbles of very dark brown partially to well-lithified mudstones, often layered or even laminated, are observed within alternating sequences of nannofossil ooze and radiolarian ooze of late to late middle Eocene age. In hand specimen, the partially lithified mudstones are particularly rich in clay and show evidence of partial secondary silicification. Pieces of porcellanite contain clay minerals, microcrystalline quartz, opaques, and calcite, as well as biogenic shells and fragments from radiolarians and foraminifers. Sediments from Sites U1331 and U1332 appear to document the silicification process in clay-rich horizons near basement, which will likely extend the findings of Moore (2008a, 2008b).

Age transect of seafloor basalt

At Site U1332 we recovered what appear to be fresh fragments of seafloor basalt, aged between 49 and 50 Ma as estimated from biostratigraphic results. This material will, when combined with other PEAT basalt samples, provide important sample material for the study of seawater alteration of basalt.

Site U1333

Three holes were cored at Site U1333 (10°30.996′N, 138°25.159′W; 4853 mbsl) (Fig. F59; see Table T1 in the "Site U1333" chapter). At Site U1333, seafloor basalt is overlain by ~183 m of pelagic sediment, dominated by nannofossil and radiolarian ooze with varying amounts of clay (Fig. F60). The oldest sediment is of early middle Eocene age.

In Hole U1333A, APC-cored sediments were recovered from ~3 m below the mudline (~4850 mbsl) to 95 m CSF (Core 320-U1333A-10H). XCB coring advanced to 184.1 m DSF through an ~60 m thick sequence of lowermost Oligocene carbonate oozes and nannofossil-bearing Eocene sediments. Near the basal section, we recovered a 30 cm long interval of lithified carbonate in Core 320-U1333A-20X. The following Core 21X contained a limestone basalt breccia. A 6 cm piece of basalt was recovered in Core 22X.

Coring in Hole U1333B started 5 m shallower than in Hole U1333A to recover the mudline and to span the core gaps from the first hole. A total of 7.73 m of carbonate-bearing ooze overlain by a few meters of clay were recovered in Core 320-U1333B-1H. Although the cores recovered from Hole U1333A showed significant porcellanite layers, we used the APC drillover strategy in Hole U1333B to obtain APC cores across and below the Eocene–Oligocene transition to 162.7 m CSF. We then XCB cored to basement and a total depth of 180.3 m CSF.

Hole U1333C was designed to provide stratigraphic overlap and confirm stratigraphic correlations made between Holes U1333A and U1333B. APC coring in Hole U1333C started 2.75 m shallower than in Hole U1333B and reached 163.2 m CSF before we had to switch to XCB coring. No downhole logging was conducted at Site U1333.

The sediment column at Site U1333 has a strong resemblance to that of Site 1218 (Lyle, Wilson, Janecek, et al., 2002) but with notably more carbonate-bearing sediments in the Eocene portion. The ~183 m of pelagic sediments has been divided into four major lithologic units (Fig. F61; see Table T2 in the "Site U1333" chapter). Unit I is ~7 m thick and contains an alternating sequence of clay, clayey radiolarian ooze, radiolarian clay, clayey nannofossil ooze, and nannofossil ooze of early Miocene age. Unit II is ~112 m thick and composed of alternating very pale brown nannofossil ooze and yellowish brown nannofossil ooze with radiolarians of early Miocene to latest Eocene age. Unit III is ~60 m thick and composed of Eocene biogenic sediments comprising clayey nannofossil ooze, nannofossil radiolarian ooze, nannofossil ooze, radiolarian nannofossil ooze, and porcellanite of latest Eocene to middle Eocene age (Unit III). Unit III is divided into two subunits, based on the absence (Subunit IIIa) or presence (Subunit IIIb) of porcellanite, which occurs between ~168 and 174 m CSF. Unit IV is a thin unit (~3.3 m) of lithified carbonate (partly limestone) and nannofossil ooze, overlying basalt (Unit V).

All major microfossil groups were found in sediments from Site U1333 and provide a consistent, coherent, and high-resolution biostratigraphic succession from basement to the top of lithologic Unit II. Shipboard biostratigraphy indicates that sediments recovered at Site U1333 span a near-continuous succession from around the lower Miocene boundary to the middle Eocene. Radiolarians are common and well preserved in the Eocene succession but less well preserved in the Oligocene sediments. A complete sequence of radiolarian zones from RN2 to RP14 (middle Eocene) was described. Initial assessment of the radiolarian assemblages across the Eocene/Oligocene boundary interval indicates a significant loss of diversity through this apparently complete succession. Although a few species from the Eocene carry through to the Oligocene, only one stratigraphic marker species (Lithocyclia angusta) first appears near the Eocene/Oligocene boundary. Calcareous nannofossils are present and moderately to well preserved through most of the succession, although there are some short barren intervals in the middle to upper Eocene. The succession spans a complete sequence of nannofossil zones from lower Miocene Zone NN1 to middle Eocene Zone NP15. The Oligocene/Miocene boundary is bracketed by the base of Sphenolithus disbelemnos in Sample 320-U1333A-2H-5, 70 cm (16.20 m CSF), and the presence of rare Sphenolithus delphix in Sample 320-U1332A-2H-CC (9.57 m CSF). Discoasters are very rare in basal assemblages, indicative of a eutrophic environment and consistent with the paleolatitude of this site in the early middle Eocene within the equatorial upwelling zone. Planktonic foraminifers are relatively abundant and well preserved from the lowest part of the Miocene to the lower Oligocene. Oligocene fauna is characterized by the common presence of Catapsydrax spp., Dentoglobigerina spp., and Paragloborotalia spp. In contrast, upper Eocene sediments contain poorly preserved specimens or are barren of planktonic foraminifers. Preservation and abundance slightly increased in some intervals of the middle Eocene, which is recognized by the presence of acarininids and clavigerinellids. The absence of the genera Globigerinatheka and Morozovella makes precise age determination of individual samples problematic. High abundances of Clavigerinella spp. have been linked to high-productivity environments, consistent with the paleogeographic location of this site (Coxall et al., 2007). Benthic foraminifers were almost continuously present and indicate lower bathyal to abyssal depths. Oligocene fauna is characterized by calcareous hyaline forms, such as Nuttallides umbonifer, Oridorsalis umbonatus, and Cibicidoides mundulus. Nuttallides truempyi and O. umbonatus often dominate the Eocene fauna. Benthic foraminifers are present through most of the section apart from an interval in the middle Eocene equivalent to radiolarian Zone RP16. They indicate lower bathyal to abyssal paleodepths.

Sedimentation rates at Site U1333 are ~6 m/m.y. in the upper sediment column from the early Miocene to the late Oligocene. In the early Oligocene, linear sedimentation rates increase to ~12 m/m.y. Between ~31 and 39 Ma (early Oligocene to early late Eocene), sedimentation rates are ~4 m/m.y. but increase slightly to ~5 m/m.y. in the interval from ~39 to 45 Ma (middle Eocene).

Paleomagnetic results from measurements made along split-core sections and on discrete samples from Site U1333 provide a well-resolved magnetostratigraphy. Shipboard analyses suggest that a useful magnetic signal is preserved in most APC-cored intervals after removal of the drilling-induced overprint by partial AF demagnetization at 20 mT. The overprint was nearly absent in those cores collected in nonmagnetic core barrels at Site U1333, whereas it was quite prominent for cores recovered in standard steel core barrels. Paleomagnetic directions from discrete samples agree well with those from split cores, confirming that AF demagnetization at 20 mT is generally sufficient to resolve the primary paleomagnetic direction regardless of which type of core barrel was used. Cleaned paleomagnetic data provide a series of distinct ~180° alternations in declination and subtle changes in inclination, which, when combined with biostratigraphic age constraints, allow a continuous magnetostratigraphy to be constructed that correlates well with the geomagnetic polarity timescale. The magnetostratigraphic record extends from the base of Chron C6n (19.722 Ma) at 1.7 m CSF in Hole U1333C to the top of Chron C20r (43.789 Ma) at 161.6 m CSF in Hole U1333C. Highlights include very high quality paleomagnetic data across Chrons C13r and C13n, which span the latest Eocene and earliest Oligocene, and a newly recognized cryptochron within Chron 18n.1n.

Geochemistry results indicate that samples from the upper part of Site U1333 have modest CaCO₃ contents of 26–69 wt% between 0 and 4 m and have frequent variations between 58 and up to 93 wt% in the interval between 4 and 35 m CSF. Calcium carbonate contents are consistently high (75.5–96 wt%) from 35 to 111 m CSF, whereas in the Eocene (between 111 and 171 m CSF) CaCO₃ contents vary abruptly between <1 and 74 wt%. The lowermost lithified carbonate rocks between 173 and 180 m CSF have high CaCO₃ contents between 76 and 90 wt%. TOC content, as determined by the acidification method, is generally very low. Pore water alkalinity values are never elevated, but alkalinity and dissolved strontium values are somewhat higher near the Eocene–Oligocene transition; these are generally consistent with carbonate dissolution or recrystallization processes. Dissolved silica increases with depth, with values always <1000 µM.

A full physical property program was run on cores from Holes U1333A–U1333C comprising WRMSL measurements of magnetic susceptibility, bulk density, and P-wave velocity; NGR; and measurements of color reflectance, followed by discrete measurements of moisture and density properties, sound velocities, and thermal conductivity on Hole U1333A cores only. All track data show variability throughout the section, allowing a detailed correlation among holes primarily using magnetic susceptibility and density (magnetic susceptibility varies around 24 × 10^–5 SI in radiolarian ooze–dominated sections and ~3 × 10^–5 SI in more carbonate rich intervals). Magnetic susceptibility values gradually increase uphole. NGR measurements are elevated by an order of magnitude in the uppermost clays and increase near the lower Oligocene at ~115 m CSF (from 5 to 8 cps). P-wave velocity gradually increases downhole as we move from carbonate- to radiolarian-dominated successions. P-wave velocity generally varies between 1490 and 1560 m/s depending on lithology, with lower velocities corresponding more to carbonate-rich sections. Bulk density and grain density show a marked decrease at ~112 m CSF (~1.70 to 1.31 g/cm³ in bulk density), where carbonate content decreases rapidly. Porosity values are generally high in the radiolarian-rich sediments (80%) and decrease in the carbonate-rich section (~60%). Thermal conductivity measurements are increased in carbonate-rich intervals and range from ~0.8 W/(m·K) in lithologic Unit I to 1.2–1.3 W/(m·K) in lithologic Unit II.

Stratigraphic correlation indicated that a complete section was recovered to ~130 m CSF in the upper Eocene, equivalent to a composite depth of ~150 m CCSF-A. For Site U1333, a growth factor of 15% is estimated from the ratio between the CCSF-A and CSF depth scales. Stratigraphic correlation with Site 1218 suggests a complete stratigraphic section in the Oligocene to uppermost Eocene interval.

Five formation temperature measurements were conducted in Hole U1333B with the APCT-3. These temperature measurements, when combined with thermal conductivity values obtained from the cores, indicate that Site U1333 has a heat flow of 42.3 mW/m² and a thermal gradient of 37.9°C/km.

Highlights

High carbonate fluctuations in middle Eocene sediments

Coring at Site U1333 was designed to capture a time interval when the CCD was slightly deeper within the middle Eocene interval that showed prominent fluctuations of carbonate content (Lyle et al., 2005). This interval occurs during the cooling that took place after the EECO (Zachos et al., 2001a) and before the Eocene–Oligocene transition (e.g., Coxall et al., 2005). Unlike Site 1218, Site U1333 sediments show carbonate contents >75 wt% in this interval at a deeper water depth and apparently coeval with the CCD cycles described by Lyle et al. (2005). Basal lithologic Unit IV recovered partially lithified carbonates.

MECO, Eocene–Oligocene and Oligocene–Miocene transitions, and depth transects

Site U1333 forms the third oldest and deepest component of the PEAT depth transect component and can be directly compared with Site 1218, which will allow the study of critical intervals (such as the Eocene–Oligocene transition; see Coxall et al., 2005) and variations of the equatorial CCD. Site U1333 is estimated to have been ~3.8 km deep during the Eocene–Oligocene transition, ~1 km shallower than today and 200 m shallower at that time than Site U1332. Carbonate content in these sediments does not change as rapidly as at the deeper and older Sites U1332 and U1333. Some of these sediments appear to be Eocene–Oligocene transition sediments that are suitable for paleoceanographic studies using carbonate-based geochemical proxies and thus are an improvement over Site 1218. Of note, Site U1333 also contains high carbonate content–bearing sediments around the MECO event (Bohaty and Zachos, 2003; Bohaty et al., 2009), allowing a detailed study of the sequence of events linking carbonate preservation cycles (Lyle et al., 2005) with climatic oscillations.

Carbonate-bearing sediments across the Oligocene–Miocene transition were also recovered at Site U1333, adding important data to the study of this time interval in the context of the PEAT Oligocene/Miocene depth transect.

Age transect of seafloor basalt

At Site U1333 we recovered what appear to be fresh fragments of seafloor basalt overlain by sediments aged 45 to 46 Ma as estimated from biostratigraphic results. This material will, when combined with other PEAT basalt samples, provide important sample material for the study of seawater alteration of basalt.

Site U1334

Three holes were cored at Site U1334 (7°59.998′N, 131°58.408′W; 4799 mbsl) (Fig. F62; see Table T1 in the "Site U1334" chapter), targeting the events bracketing the Eocene–Oligocene transition as part of an investigation of the wider Cenozoic climatic evolution (e.g., Zachos et al., 2001a) and providing data toward a depth transect across the Oligocene (see "Eocene/Oligocene Boundary [Site U1334; 38 Ma crust]") that will allow exploitation and verification of a previous astronomical age calibration from Site 1218 (Pälike et al., 2006a).

Site U1334 is in the center of the PEAT program transect, ~100 km north of the Clipperton Fracture Zone and ~380 km southeast of the previously drilled Site 1218. At Site U1334, seafloor basalt is overlain by ~285 m of pelagic sediment. The oldest sediment is of late middle Eocene age (38 Ma).

The topmost ~47 m thick lithologic Unit I contains a 15 m thick interval of brown radiolarian clay overlying ~32 m of alternating radiolarian clay and nannofossil ooze. The uppermost section (320-U1334A-1H-CC) is of late Miocene age (radiolarian Zone RN7; ~8.5 Ma). Below, Unit II comprises a ~200 m thick succession of upper Miocene to Oligocene nannofossil ooze and chalk above a ~35 m thick sequence of upper Eocene nannofossil chalk, radiolarite, and claystone (Unit III). Basal lithologic Unit IV (~1 m thick; 285 m CSF) consists of middle Eocene intercalated micritic chalk and limestone on basalt (Figs. F63, F64).

Holes U1334A–U1334C provided high-quality APC-cored sediments from the mudline to ~210 m CSF (Cores 320-U1334A-22H, 320-U1334B-22H, and 320-U1334C-22H). Below this depth we encountered increasingly stiffer and harder sediment, after which we switched to the XCB cutting shoe. XCB coring advanced to 288.5 m DSF through lower Oligocene and Eocene sediments with high recovery. At the base of the holes, an intercalated unit of basalt and hard micritic chalk and limestone occurs below a 10–20 m thick section of nannofossil ooze and chalk. For detailed coring activities, see "Operations" in the "Site U1334" chapter.

Carbonate content exceeds 92 wt% in the upper lower Miocene below Section 320-U1334A-5H-3 and remains high throughout the Oligocene. Eocene sediments still contain considerable amounts of carbonate, and nannofossil ooze and chalk are dominant lithologies apart from several short less carbonate rich intervals (e.g., Section 320-U1334A-28X-3). In the middle Eocene, carbonate content cycles between ~40 and 85 wt% (see "Geochemistry" in the "Site U1334" chapter) (Fig. F65), with higher values encountered toward the basal part of the Eocene section. Two short intervals in the upper Eocene (~249 to ~257 m CSF) exhibit carbonate content of <20 wt%.

A series of middle Oligocene cores (Cores 320-U1334A-16H through 21H) were recovered that had very distinct colors ranging from light grayish green to light blue (see "Lithostratigraphy" in the "Site U1334" chapter). These uniquely colored carbonate oozes exhibit extremely low magnetic susceptibilities that complicated stratigraphic correlation. These oozes have lost almost their entire magnetic susceptibility signal from ~145 to ~215 m CSF (Figs. F64, F66). Similar colored cores have previously been described for Sites 78 and 79 (Hays et al., 1972).

The Eocene–Oligocene transition at Site U1334 is much more expanded than at IODP Sites U1331–U1333 and even Site 1218. The transition was encountered at ~250 m CSF and fully recovered in Cores 320-U1334A-27X and 320-U1334B-26X; Hole U1334C was used to fill small stratigraphic gaps. The Oligocene–Miocene transition was fully recovered in all three holes in Cores 320-U1334A-10H (based on magnetostratigraphy, the boundary is at Sample 320-U1334A-10H-6, 98 cm), 320-U1334B-10H (top of Section 2), and 320-U1334C-10H.

All major microfossil groups occur in sediments from Site U1334 and provide a consistent, coherent, and high-resolution biostratigraphic succession spanning a near-continuous sequence from the middle Miocene to the uppermost middle Eocene. The uppermost 12 m of radiolarian clay is barren of calcareous microfossils but contains radiolarians of middle Miocene age, similar to the site survey piston Core RR0306-08JC (Lyle et al., 2006). Nannofossil ooze and radiolarian clays are present in the Miocene and Eocene parts of the section, with nannofossil ooze dominant in the thick Oligocene section. Radiolarians are present through most of the section, apart from the lowermost cores, and are well preserved in the Eocene. They provide a coherent high-resolution biochronology and indicate a complete sequence of radiolarian zones from RN7 (upper Miocene) to RP17 (uppermost middle Eocene). Calcareous nannofossils are present and moderately to well preserved through most of the succession, and there appears to be a complete sequence of nannofossil zones from NN6 (middle Miocene) to NP17 (uppermost middle Eocene), providing a minimum age estimate for basaltic basement of 38 Ma. In the Eocene, the base of Chiasmolithus oamaruensis is determined in Sample 320-U1334A-30X-1, 66 cm, and the top of Chiasmolithus grandis in Sample 320-U1334-30X-2, 74 cm. Intriguingly, both species are mid- to high-latitude taxa (Wei and Wise, 1989) and are present only rarely and sporadically at Site U1334. Planktonic foraminifers are present through most of the succession and are relatively abundant and well preserved from the lower Miocene to the lower Oligocene. The lower Miocene is characterized by the presence of Dentoglobigerina spp., Paragloborotalia siakensis–mayeri, Paragloborotalia kugleri, and Paragloborotalia pseudokugleri. Oligocene sediments contain Catapsydrax spp., Paragloborotalia opima-nana, and characteristic Dentoglobigerina spp. Preservation and abundance of planktonic foraminifers is more variable in the middle Miocene and upper Eocene/lowermost Oligocene. No Eocene/Oligocene boundary marker hantkeninids were identified. Benthic foraminifers are present through most of the section and indicate lower bathyal to abyssal paleodepths.

Sedimentation rates, as derived from magneto- and biostratigraphic age determinations, vary throughout the section and are ~4 m/m.y. in the topmost sediment cover, vary between ~12 and 14 m/m.y. in the lower Miocene through upper lower Oligocene section, increase to ~24 m/m.y. in the lower Oligocene, and are ~8 m/m.y. in the upper Eocene. There is no obvious hiatus in the shipboard biostratigraphic sequence. The presence of all major fossil groups as well as a detailed and well-resolved magnetostratigraphy will allow us to achieve one of the main PEAT objectives of arriving at an integrated Cenozoic stratigraphy and age calibration for major parts of the Miocene, Oligocene, and Eocene.

A full physical property program was run on cores from Site U1334. This program comprises WRMSL measurements of magnetic susceptibility, bulk density, and P-wave velocity, along with NGR and measurements of color reflectance, followed by discrete measurements of moisture and density (MAD) properties, sound velocities, and thermal conductivity in Hole U1334A. All track data are variable throughout the section, allowing a detailed correlation between different holes, with the exception of a very low magnetic susceptibility signal within an interval extending slightly above and below the light greenish gray tinted cores of Unit II (see "Lithostratigraphy" in the "Site U1334" chapter for exact color definitions), between ~140 and 205 m CSF. Magnetic susceptibility varies between 10 × 10^–5 and 40 × 10^–5 SI in Unit I, oscillates around 5 × 10^–5 to 10 × 10^–5 SI above the colored sediments, and then drops to near zero and negative values, returning to values around 10 × 10^–5 SI in the lower part of Unit II and Subunit IIIa. NGR increases slightly at the Eocene/Oligocene boundary at ~246 m CSF (from 4 to 7 cps). P-wave velocity remains uniform through the upper 150 m of sediment (varying around 1500 m/s) but increases rapidly below the ooze/chalk boundary to ~1600 m/s. This explains the slightly thicker sediment section than was expected from seismic data prior to coring (~20 m thicker). For Hole U1334B, no P-wave velocity WRMSL data were collected between ~125 and 240 m CSF to allow for a more timely stratigraphic correlation of cores within the iron reduction–dominated colored cores with the GRA instrument. Bulk density and grain density increase gradually with carbonate content to ~204 m CSF to a maximum of ~1.8 g/cm³ and then show stepped decreases in the lower part of this succession. Ephemeral whole-round samples were collected at ~50 and ~165 m for shore-based studies of sediment permeability.

WRMSL data were used to achieve stratigraphic correlation among holes at Site U1334. Magnetic susceptibility was initially the main parameter used for real-time correlation, as a second loop of the susceptibility meter is mounted on the Special Task Multisensor Logger (STMSL); the second bulk density instrument on this track was not working. In the very low (sometimes negative) magnetic susceptibility interval between ~140 and ~205 m CSF (Cores 320-U1334A-16H through 21H), the signal was not useful for correlation, and we measured the corresponding cores from Hole U1334B out of sequence to establish the amount of core overlap using bulk density. The coring effort in Hole U1334C was successful at covering gaps between cores at this site to ~111 m CCSF-A, as well as from 250 to 335 m CCSF-A, almost to the bottom of the section. The correlation was challenging between the three holes at Site U1334 in the greenish–light gray interval (Cores 320-U1334A-15H through 22H, 320-U1334B-14B through 22H, and 320-U1334C-14H through 22H) and in the bottom 80 m of the section, where XCB coring compromised the GRA density variations that would otherwise help stratigraphic correlation. Visual inspection, comparison with core imagery, and biostratigraphic datums were used to establish and verify hole to hole correlation where track data lacked clearly identifiable features. Stratigraphic correlation between individual holes indicates a growth factor (ratio between the CCSF-A and CSF depth scales) of ~16%. Stratigraphic correlation resulted in a complete splice through the Eocene–Oligocene transition almost to basement (~38 Ma).

A full range of paleomagnetic analyses was conducted on 66 APC cores and 188 discrete samples from Site U1334 for the APC-cored section of Site U1334 (upper ~209 m). Unlike Sites U1331 and U1332, the drilling overprint was generally weak for Site U1334 cores, but only for those collected with nonmagnetic core barrels (Cores 320-U1334A-1H through 16H, 320-U1334B-1H through 15H, and 320-U1334C-1H through 15H). In contrast, those cores collected with steel core barrels are highly overprinted. The overprint is so severe that even demagnetization at 20 mT only partially removes it. This extreme overprint notably degrades the paleomagnetic declination data, as can be noted by their higher variability, which makes polarity determination much more difficult in the intervals collected with steel core barrels. The problem is exacerbated by the decay in the intensity (and magnetic susceptibility) that occurs at ~135 m CSF in all three holes as a result of reduction diagenesis. Even within the highly reduced interval, an interpretable signal was present prior to switching to steel core barrels. Magnetic susceptibility in the upper 45 m of Hole U1334A averages ~18 × 10^–5 SI (volume normalized) and decreases to a mean of 6 × 10^–5 SI from 45 to 135 m CSF. A notable low occurs from ~140 to 204 m CSF, where the average susceptibility is 0.6 × 10^–5 SI. This low interval is associated with a change in sediment color from yellowish tan to very light green, blue, and gray at ~143 m CSF and another abrupt change to reddish brown tones at ~192 m CSF, which corresponds to middle early Oligocene (~30 Ma). Just below 205 m, magnetic susceptibility steps up to an average of 5 × 10^–5 SI and then increases again across the Eocene/Oligocene boundary (~245 m) to an average of 18 × 10^–5 SI. The magnetostratigraphy in Hole U1334A has been interpreted from the top of Chron 11r (29.957 Ma), which occurs ~55 cm below the top of Section 320-U1334C-21H-4 (~195 m CSF) through the base of Chron C3n.4n (5.235 Ma) in Core 320-U1334A-1H (~4 m CSF). The youngest sediments recovered are in the upper ~2 m of Core 320-U1334A-1H, which record Chrons C1n through C2r.1r.

A standard shipboard suite of geochemical analyses of pore water and organic and inorganic sediment properties was undertaken on samples from Site U1334. We also conducted a high-resolution (one per section) Rhizon pore water investigation across the interval's middle Oligocene cores (320-U1334C-16H through 21H) that exhibited colored sediments. Site U1334A is marked by alkalinities between 3 and 4 mM throughout. The most striking features in the interstitial water geochemistry are a dissolved manganese peak from ~20 to ~240 m CSF with a maximum of ~6 µM at ~110 m CSF and a dissolved iron peak as high as >15 µM centered at 165 m CSF. The depth range of the dissolved iron peak, indicative of iron oxide reduction, coincides with the colorful interval seen in the lithology and with the interval of low magnetic susceptibilities (~140–205 m CSF). Sulfate results indicate limited sulfate reduction. Calcium carbonate contents are low in the uppermost ~35 m of Site U1334, and calcium carbonate contents are generally high (~80 wt%) below the uppermost clay layer.

Wireline logging was attempted in Hole U1334C with a redesigned tool string configuration after the loss of equipment at Site U1332. However, this attempt had to be abandoned after the logging winch failed when the tool was on its way down the drill pipe.

Five downhole temperature measurements were conducted in Hole U1334B with the APCT-3 tool and reveal a thermal gradient of 33°C/km. Seafloor temperature is ~1.5°C. Temperature data combined with whole-round core temperature conductivity measurements indicate the heat flow is 31.6 mW/m² at this site. This is somewhat lower than values obtained for the nearest site (1218).

Highlights

Eocene–Oligocene and Oligocene–Miocene transitions and depth transects

Site U1334 was planned as the youngest and shallowest component of the PEAT Eocene–Oligocene depth transect component, which will allow the study of critical intervals (such as the Eocene–Oligocene transition; see Coxall et al., 2005) and variations of the equatorial CCD. Site U1334 is estimated to have been ~3.5 km deep during the Eocene–Oligocene transition, ~1.3 km shallower than today and 800 m shallower at that time than Site U1333. Unlike previously drilled sites, the dominant lithology below the Eocene–Oligocene transition is still nannofossil ooze and chalk, with significant amounts of carbonate present. These carbonate amounts will allow us to achieve the prime objective for this site. The Eocene–Oligocene transition, which was cored multiple times at Site U1334, has higher sedimentation rates than previously cored examples. The overlying Oligocene is also much more expanded than at Site 1218, with better preservation of planktonic foraminifers over a longer time interval, permitting a more detailed study of the Oligocene climate system. Site U1334 contains carbonate-bearing sediments across the Oligocene–Miocene transition. Physical property data from Site U1334 can be correlated cycle by cycle to Site 1218, allowing correlation to a previously astronomically calibrated site for the Oligocene.

Geochemical front

At Site U1334 we recovered a ~50 m thick interval of light greenish gray carbonates that show a distinct peak in dissolved Fe concentrations, characteristic of a geochemical alteration front. A similar but much thicker alteration zone is also observed at IODP Site U1335 and provides the opportunity to study organic matter degradation while these sites migrate from south to north through the equatorial belts of high productivity.

Age transect of seafloor basalt

At Site U1334 we recovered what appear to be fresh fragments of seafloor basalt, aged ~38 Ma as estimated from biostratigraphic results. This material will, when combined with other PEAT basalt samples, provide important sample material for the study of seawater alteration of basalt.

Site U1335

Two holes were cored at Site U1335 (5°18.735′N, 126°17.002′W; 4327.5 mbsl) (Fig. F67; see Table T1 in the "Site U1335" chapter), targeting paleoceanographic events in the late Oligocene and into the early and middle Miocene, including and focusing on the climatically significant Oligocene–Miocene transition and the recovery from the Mi-1 glaciation event (Zachos et al., 2001b; Pälike et al., 2006a) and the expansion of the East Antarctic cryosphere (Holbourn et al., 2005). Site U1335 also provides data toward a depth transect across the latest Oligocene and Miocene (see "Latest Oligocene–earliest Miocene [Site U1335; 26 Ma crust]") that will allow exploitation and verification of a previous astronomical age calibration from Site 1218 (Pälike et al., 2006b).

Site U1335 (~26 Ma crust) is situated halfway between Site U1336 ~340 km toward the northwest and Site U1337 ~390 km toward the southeast, ~250 km south of the Clipperton Fracture Zone (Lyle et al., 2006). At Site U1335, seafloor basalt is overlain by ~414 m of pelagic sediment. The oldest sediment is of late Oligocene age (26 Ma).

The sedimentary sequence at Site U1335 is divided into two major lithologic units (see "Lithostratigraphy" in the "Site U1335" chapter). The topmost ~64 m thick lithologic Unit I comprises an alternating sequence of earliest late Miocene to Pleistocene calcareous nannofossil, diatom, radiolarian, and foraminifer oozes. The topmost sediment of Unit I is younger than the Pleistocene/Pliocene boundary (see "Biostratigraphy" in the "Site U1335" chapter) as recognized by the top of planktonic foraminifer Globigerinoides fistulosus (between Samples 320-U1335A-1H-CC and 2H-2, 104–106 cm) and then follows a continuous biostratigraphic succession to the early late Miocene. Below, lithologic Unit II comprises a ~350 m thick succession of late Miocene to late Oligocene (calcareous nannofossil Zone NP25) nannofossil ooze and chalk overlying basalt (lithologic Unit III) (Figs. F68, F69). One of the prominent features of Unit II is the presence of at least 49 described beds (2–176 cm thick) of nannofossil foraminifer ooze that have sharp basal boundaries, many of which are irregular and some of which are inclined. These beds are interpreted as gravity flow deposits from the nearby seamounts and represent ~2% of the total sediment recovered.

Holes U1335A and U1335B provided high-quality APC-cored sediments from the mudline to ~341 and 378 m CSF, respectively (Cores 320-U1335A-36H and 320-U1335B-41H). At the time it was recovered, the APC-cored interval from Hole U1335B represented the second deepest APC-cored depth in ODP and IODP history. Below this depth we encountered stiffer and harder sediment, after which we switched to the XCB cutting shoe. XCB coring advanced to ~420 m DSF through lower Miocene and upper Oligocene sediments with high recovery. In the basal section, Core 320-U1335B-46X recovered pieces of basalt up to 10 cm in length with a glassy rim and overlain by nannofossil chalks of Unit II. For detailed coring activities, see "Operations" in the "Site U1335" chapter.

The sediment column at Site U1335 represents the youngest end-member drilled during Expedition 320 and provides one of the most stratigraphically complete and expanded lower Miocene sections from the equatorial Pacific to date (~320 m cored depth from the lowermost to uppermost Miocene).

At Site U1335, carbonate content fluctuates between 12 and 87 wt% within Unit I (see "Geochemistry" in the "Site U1335" chapter) (Fig. F70), presumably reflecting the close proximity of the seafloor to the lysocline. With the exception of the depth interval from 140 to 220 m CSF, the remainder of Unit II exhibits uniformly high calcium carbonate content between 80 and 90 wt%. From ~150 to 210 m CSF (approximately equivalent to Cores 320-U1335A-16H through 22H), carbonate content cycles between ~50 and 90 wt% and corresponds to a change in dominant sediment color from light greenish gray to tan, displaying higher magnetic susceptibility values up to 25 × 10^–5 SI.

A series of upper Oligocene through upper middle Miocene cores (320-U1335A-8H through 40X) were recovered with distinct colors ranging from light grayish green to light blue (see "Lithostratigraphy" in the "Site U1335" chapter), similar but much thicker in total stratigraphic thickness (~70–170 and ~200–350 m) than those observed at Site U1334 (see the "Site U1334" chapter). The colored carbonate oozes have extremely low magnetic susceptibilities that complicated a confident stratigraphic correlation. These colored oozes have lost almost their entire magnetic susceptibility signal from ~70 to ~105 m CSF and below ~210 m CSF (Figs. F69, F71, F72). Similar colored cores have previously been described for Sites 78 and 79 (Hays et al., 1972).

All major microfossil groups occur in sediments from Site U1335, representing a complete biostratigraphic succession at the shipboard sample resolution level of Pleistocene to uppermost Oligocene sediments, including a thick sequence of lower Miocene nannofossil ooze and chalk (see "Biostratigraphy" in the "Site U1335" chapter). Radiolarians are present through most of the section apart from the basal 3 m of nannofossil chalk. They provide a coherent high-resolution biochronology through a complete sequence of radiolarian zones from RN14 (Pleistocene) to RP21 (upper Oligocene). Calcareous nannofossils are present and moderately to well preserved through most of the succession, representing the complete sequence from Zone NP25 (upper Oligocene) above basaltic basement through Zone NN20 (Pleistocene). Planktonic foraminifers are present throughout the succession and are moderately to well preserved. Recognized planktonic foraminifer zones range from Zone PT1a (Pleistocene) to Zone O6 (upper Oligocene). Nannofossil, radiolarian, and planktonic foraminifer datums are in good agreement. Benthic foraminifers are present through most of the section and indicate lower bathyal to abyssal paleodepths. The Oligocene–Miocene transition at Site U1335 was encountered at ~350 m and was fully recovered in Cores 320-U1335A-37X and 320-U1335B-38H as approximated by the planktonic foraminifer datum base of Paragloborotalia kugleri between Samples 320-U1335A-37X-4, 136–138 cm, and 37X-CC (midpoint = 348.6 m CSF), in good agreement with the calcareous nannofossil event top of Sphenolithus delphix at 349.7 m CSF between Samples 320-U1335A-37X-6, 50 cm, and 37X-CC. The oldest sediment overlying seafloor basalt has been assigned to calcareous nannofossil Zone NP25 (24.4–26.8 Ma).

Sedimentation rates, as derived from the magneto- and biostratigraphic age determinations (see "Stratigraphic correlation and composite section" in the "Site U1335" chapter), vary throughout the section and are ~6 m/m.y. in the late to middle Miocene to recent sediment cover, ~17 m/m.y. in the middle early Miocene and as high as ~25 m/m.y. throughout the late Oligocene and early Miocene. There is no obvious hiatus at the shipboard biostratigraphic resolution, although some condensed horizons exist (e.g., near the early/middle Miocene boundary and in the early late Miocene; see "Biostratigraphy" in the "Site U1335" chapter). The presence of all major fossil groups as well as a detailed and partly well resolved magnetostratigraphy will allow us to achieve one of the main PEAT objectives of arriving at an integrated Cenozoic stratigraphy and age calibration for the Miocene and late Oligocene.

A full physical property program was run on cores from Site U1335. This program comprises WRMSL measurements of magnetic susceptibility, bulk density, and P-wave velocity; NGR; and measurements of color reflectance, followed by discrete measurements of MAD properties, sound velocities, and thermal conductivity in Hole U1335A. All track data vary throughout the section, allowing a detailed correlation among holes, with the exception of a low magnetic susceptibility signal within an interval extending slightly above and below the light greenish gray tinted cores of Unit II (see "Lithostratigraphy" in the "Site U1335" chapter for exact color definitions), between ~70 and 110 and ~210 and ~380 m CSF. Magnetic susceptibility varies between 5 × 10^–5 and 20 × 10^–5 SI in the upper parts of Unit I and then increases to ~25 × 10^–5 SI toward the lower part of Unit I, coinciding with the presence of clayey radiolarian ooze within the major lithology of nannofossil ooze. Magnetic susceptibility values decrease at the top of Unit II (~64 m CSF) and then fall to values around –1 × 10^–5 SI near 70 m CSF. Between ~110 and 150 m CSF, magnetic susceptibility values increase slightly and become highly variable (0 to 10 × 10^–5 SI). Magnetic susceptibility values are higher in the interval from 160 to 200 m CSF, coinciding with an observed decrease in Fe reduction (see "Lithostratigraphy" in the "Site U1335" chapter). Below 200 m CSF, the magnetic susceptibility signature is dominantly diamagnetic, with values close to zero. Magnetic susceptibility values slightly increase again in the basal 20 m of Unit II (below ~400 m CSF). NGR is elevated at the surface sediment (~73 cps) but low throughout the rest of the sedimentary column. P-wave velocities from the WRMSL agree with discrete velocity measurements and reflect key lithologic transitions, particularly the ooze to chalk transition near ~220 m CSF. P-wave velocities are between 1460 and 1490 m/s in Unit I and the upper part of Unit II and then increase to >1500 m/s. Slightly below the ooze–chalk transition near 345 m CSF, velocities increase significantly, reaching 1600–1750 m/s at the bottom of Unit II. This partly explains the thicker sediment section than was expected from seismic data prior to coring (~60 m thicker). Bulk density and grain density increase with depth, with an increase in wet bulk density from 1.2 to 1.6 g/cm³ in Unit I to ~1.7 g/cm³ at the top of Unit II and ~1.8 g/cm³ in the basal part of the section. Sediment porosity ranges from 70% to 90% in Unit I to 50%–60% at ~300 m CSF in Unit II. Ephemeral whole-round samples were collected at ~96, ~196, and ~305 m CSF for shore-based studies of sediment permeability.

The coring effort in Holes U1335A and U1335B was successful at covering stratigraphic gaps between cores at this site from the surface throughout most of the APC-cored section (see "Stratigraphic correlation and composite section" in the "Site U1335" chapter), with the exception of a gap (~1 m) at the bottom of Core 320-U1335A-16H due to flow-in (~146.40–151.46 m CSF). Features in magnetic susceptibility and GRA density are well aligned down to a depth of 337 m CSF (Hole 1335A) and 344 m CSF (Hole U1335B), corresponding to ~398 m CCSF-A. Between ~230 and ~398 m CCSF-A, GRA density data allowed confident alignment of cores despite very low magnetic susceptibility values. The section below ~398 m CCSF-A was mostly XCB cores, lacked clearly identifiable features, and therefore had to be appended to the splice. A single spliced record was assembled for the aligned cores down to Section 320-U1335B-37H-6 (343.76 m CSF; 398.15 m CCSF-A). Stratigraphic correlation between individual holes indicates a growth factor (ratio between the CCSF-A and CSF depth scales) of ~16%. Stratigraphic correlation resulted in a complete splice through the Eocene–Oligocene transition almost to basement (~38 Ma).

A full range of paleomagnetic analyses was conducted on 78 archive halves and 257 discrete samples from Site U1335 for the APC-cored section (upper ~378 m). The most prominent feature of the records is the magnetic intensity and susceptibility low that occurs between ~70 and 110 m CSF and below ~210 m CSF. We could not obtain any reliable paleomagnetic directions from this interval because the magnetic intensity after 20 mT AF demagnetization is on the order of 10^–5 A/m, which is comparable to the noise level of the superconducting rock magnetometer (see "Paleomagnetism" in the "Site U1335" chapter). The drilling overprint was generally weak when nonmagnetic core barrels were used (Cores 320-U1335A-1H through 16H and 320-U1335B-1H through 19H). In contrast, cores collected with the steel core barrels are highly overprinted. Except for the low magnetic intensity interval, the cleaned paleomagnetic data provide a series of distinct ~180° alternations in declination. When combined with biostratigraphic age constraints (see "Biostratigraphy" in the "Site U1335" chapter), the data allow a continuous magnetostratigraphy from Chrons C1n (0–0.781 Ma) to C5n.2n (9.987–11.040 Ma) from 0 to 65.95 m CSF in Hole U1335A and from Chrons C1n to C5r.1n (11.118–11.154 Ma) from 0 to 66.225 m CSF in Hole U1335B. Below the bottom of the first magnetic low zone (~70–110 m CSF), magnetostratigraphy is again interpretable downhole: from Chrons C5Br (15.160–15.974 Ma) to C6n (18.748–19.722 Ma) from 155.35 to 208.40 m CSF in Hole U1335A and from Chrons C5AAn (13.015–13.183 Ma) to C5Er (18.524–18.748 Ma) from 107.95 to 202.60 m CSF in Hole U1335B. The highlights of the magnetostratigraphy at Site U1335 are the identifications of (1) a previously observed cryptochron (C5Dr-1n) in two holes and (2) 40 potential geomagnetic excursions (10 of which are recorded in both holes).

A standard shipboard suite of geochemical analyses of pore water and organic and inorganic sediment properties was undertaken on samples from Site U1335. Site U1335 is marked by alkalinities between 2.5 and 4.3 mM throughout, sulfate concentrations between 23 and 28 mM, and dissolved phosphate concentrations of ~2 µM in the shallowest sample, decreasing to ~0.5 µM in the uppermost ~50 m. The most striking features in the interstitial water geochemistry are three dissolved manganese peaks with concentrations of up to 44, 13, and 5 µM at ~0–40, 50–80, and 150–210 m CSF, respectively. Dissolved iron also shows three peaks, with concentrations up to 6 µM at ~6 m CSF, between 90 and 170 m CSF, and between 190 and 370 m CSF. Minima in dissolved Fe correspond to elevated Mn concentrations. The alternating pattern of dissolved Mn and Fe correspond well to apparent color changes in the sediment column (see "Lithostratigraphy" in the "Site U1335" chapter). Lithium concentrations decrease from ~26 µM at the sediment surface to 5 µM at ~300 m CSF, below which Li concentrations increase strongly to ~32 µM. The Sr concentration profile mirrors that of Li, with concentrations ranging between 82 and 250 µM. Sr values increase from the top to 200 m CSF, followed by a decrease toward basement. Calcium carbonate, IC, and TC contents were determined on sediment samples from Hole U1335A (Fig. F70). CaCO₃ contents ranged between 13 and 96 wt%. In the uppermost ~67 m, carbonate contents range from 12 to 87 wt%, and concentrations are then consistently high (~72–96 wt%) between 67 and 157 m CSF and below 222 m CSF. Carbonate contents vary more widely (between 37 and 89 wt%) from 157 to 222 m CSF. TOC concentrations were determined by acidification and are generally low.

Wireline logging was not conducted at Site U1335. Five downhole temperature measurements were conducted in Hole U1335B with the APCT-3 and reveal a thermal gradient of 7.5°C/km. Temperature data combined with whole-round core temperature conductivity measurements indicate the heat flow is 7 mW/m² at this site. This is much lower than values obtained for any of the other Expedition 320 sites and would suggest recirculation of seawater through basement, consistent with some of the interstitial pore water results (see "Geochemistry" in the "Site U1335" chapter).

Highlights

Highly expanded Miocene sedimentary section

One of the highlights from Site U1335 is the recovery of a thick Miocene carbonate-dominated section from the central equatorial Pacific, one of the high-priority objectives of the PEAT program. The early Miocene (7.1 m.y. duration) is captured in ~190 m of sediment, corresponding to a sedimentation rate of 27 m/m.y. The middle Miocene (4.4 m.y. duration) is recovered in ~95 m sediment, with a sedimentation rate of ~21 m/m.y. The sedimentation rate from the late Oligocene into the Miocene is just under 20 m/m.y. These high sedimentation rates will facilitate the study of paleoceanographic processes at unprecedented resolution for the equatorial Pacific.

Oligocene–Miocene transition and depth transects

Site U1335 was planned as the youngest and shallowest component of the PEAT Oligocene–Miocene depth transect component, which will allow the study of critical intervals (such as the Mi-1 glacial inception; see Zachos et al., 2001b; Pälike et al., 2006a) and variations of the equatorial CCD throughout this transition and during the latest Oligocene and early Miocene. Site U1335 is estimated to have been ~3.3 km deep during the Oligocene–Miocene transition, ~1.5 km shallower than today. The dominant lithologies are nannofossil ooze and chalk, with better preservation of calcareous microfossils than any other site drilled during Expedition 320, which will allow us to achieve the prime objective for this site. Physical property data from Site U1335 provide an important contribution toward the Cenozoic megasplice, connecting with younger sediments from ODP Leg 138 (e.g., Site 850) and older sediments from Leg 199 (Site 1218), allowing the generation of astronomically calibrated datums and isotope stratigraphies from the Miocene into the Eocene.

Geochemical front

At Site U1335 we recovered an interval of light greenish gray carbonates that show a distinct peak in dissolved Fe concentrations, characteristic of a geochemical alteration front. At Site U1335, this zone is similar to but much thicker in total stratigraphic thickness (~70–170 and ~200–350 m CSF) than that observed at Site U1334 (~50 m; see the "Site U1334" chapter). Although the paleomagnetic signal was lost in most parts of this section, sediments recovered will provide the opportunity to study organic matter degradation while these sites migrated from south to north through the equatorial belts of high productivity. Paleolatitudinal reconstructions show that these characteristic geochemical alteration fronts can be mapped to similar equatorial positions between Sites U1334 and U1335, roughly between the Equator and ~2°N. One feature of interest at Site U1335 is the observation that the multicolored interval of sediments is interrupted between ~170 and 200 m CSF (Cores 320-U1335A-18H through 20H), again showing higher magnetic susceptibility values. It remains to be established whether this interruption in the geochemical alteration front is related to the shape and position of the equatorial high-productivity zone or instead is the result of reduced sedimentation rates during this time (late early Miocene). Interstitial pore water profiles provide additional important information about the redox chemical processes operating in this zone (see "Geochemistry" in the "Site U1335" chapter), which have also been observed at Sites 78, 79, and 574 (e.g., Hays et al., 1972).

Gravity flow deposits

One of the prominent features of Unit II is the presence of at least 49 described beds (2–176 cm thick) of nannofossil foraminifer ooze that have sharp basal boundaries, many of which are irregular and some of which are inclined. These beds are interpreted as gravity flow deposits from the nearby seamounts and represent ~2% of the total sediment recovered. Their grain size fines upward from medium sand to silt, and they are often darker colored than immediately overlying deposits and instantly recognizable by their coarser texture. Angular basalt fragments (<1 mm), fish teeth, and pyritized foraminifers and radiolarians were also found within the basal parts of these beds, of which at least three show parallel or cross-laminations in their upper or middle part. These beds, interpreted as gravity flow deposits, are present with an approximate frequency of one or two beds per core. The abundance and thickness of these beds is highest within Cores 320-U1335-21H through 37X (189.4–350.1 m CSF). No gravity flow deposits were observed in Cores 320-U1335A-3H through 8H. The provenance of these deposits, as indicated by the observed basalt fragments, is inferred to be the nearby seamounts (Fig. F67B) situated ~15–20 km northeast and southeast of Site U1335, with present summit water depths that are 400–600 m shallower than Site U1335. Initial indications are that these gravity flow deposits, unlike those observed at Site U1331, might not be very erosive and therefore essentially add to the sediment column rather than removing large sections of geological time. The high sedimentation rates at Site U1335 will allow paleoceanographic studies to avoid the generally thin layers of gravity flows.

Age transect of seafloor basalt

At Site U1335 we recovered what appear to be fresh fragments of seafloor basalt with an age of ~26 Ma, as inferred by the oldest biostratigraphic datums from the sediment above. This material will, when combined with other PEAT basalt samples, provide important sample material for the study of seawater alteration of basalt.

Site U1336

Two holes were cored at Site U1336 (proposed Site PEAT-5C; 7°42.067′N, 128°15.253′W; 4286 mbsl) (Fig. F73; see Table T1 in the "Site U1336" chapter) targeting paleoceanographic events in the late Oligocene and into the Miocene, including a focus on the Oligocene–Miocene transition and the recovery of the Mi-1 glaciation event (Zachos et al., 2001b; Pälike et al., 2006b). In conjunction with Sites U1335 and U1337, Site U1336 was also designed to provide a latitudinal transect for early Miocene age slices. Site U1336 provides data toward a depth transect across the late Oligocene and Miocene that allow us to verify and apply a previous astronomical age calibration from Site 1218 (Pälike et al., 2006a).

At Site U1336, APC cores were taken from the seafloor to 184.8 m (Cores 320-U1336A-1H through 21H) and 173.6 m (Cores 320-U1336B-1H through 20H). Nonmagnetic core barrels were used for Cores 320-U1336A-1H through 16H and Cores 320-U1336B-1H through 16H and steel barrels were used for all other cores. Two hard layers, one at ~121 m CSF (Cores 320-U1336A-14H and 320-U1336B-14H) and one at ~135 m CSF (Core 320-U1336B-16H) caused core loss and prevented the development of a continuous sediment section. XCB cores (320-U1336A-22X through 35X) were taken from 184.8 to 302.9 m CSF in Hole U1336A. We stopped coring before reaching the basement objective because of decreasing rates of penetration, relatively low recovery, and the possibility of obtaining a stratigraphically complete Miocene section by allocating the remaining operational time during Expedition 320 to Hole U1336B.

At Site U1336, ~300 m of pelagic sediments are divided into three major lithologic units (Fig. F74). The sediments are composed mainly of nannofossil oozes, nannofossil chalks, and chert. The lower to middle Miocene sedimentary sequence of Unit I (0–74.54 m CSF) contains more radiolarians, clay, foraminifers, and diatoms relative to the lower Miocene to lower Oligocene sediments below ~70 m CSF. Subtle changes in the relative proportions of these minor components produce meter-scale dark–light color cycles and two diatom-rich layers. Numerous rounded fragments of pumice occur throughout this unit.

Unit II (74.50–189.50 m CSF) is dominated by nannofossil ooze. Sediment color changes downhole from pale yellow to light greenish gray at 92 m CSF. Below this boundary, the color of Unit II alternates between light greenish gray and white to 184.80 m CSF. Oxidation-reduction reactions are responsible for the observed vivid colors and pore water chemistry changes, likely fueled by varied availability of organic carbon. Occasional thin chert layers were encountered below 120 m CSF in Unit II. Mainly broken chert fragments were recovered, except for a small in situ chert fragment at 159.6 m CSF in Section 320-U1336B-18H-4, 106 cm. More abundant chert layers are common in the lower third of the recovered sequence.

Unit III (189.5–299.6 m CSF) was only recovered in Hole U1336A. The dominant lithologies of this unit are light greenish gray and white nannofossil chalk with light greenish gray millimeter-scale color banding and chert layers. The chert shows many different colors including black, dark greenish gray, very dark greenish gray, dark gray, olive-yellow, dark brown, and pink. The Unit II–III transition is identified by the uppermost common occurrence of chert. Below 289 m CSF, nannofossil chalk contains increasing amounts of micrite and the cherts vary in color. The lowermost cherts are olive-yellow, then pink, and, finally, dark brown at the base. The chalk changes color to white below 298.54 m CSF. CaCO₃ contents remain >88 wt% in the chalk layers. Igneous basement was not recovered at Site U1336.

All major microfossil groups were found in sediments from Site U1336, representing a complete biostratigraphic succession at the shipboard sample resolution level of middle Miocene to lower Oligocene sediments. They provide a coherent, high-resolution biochronology through a complete sequence (Fig. F74). Calcareous nannofossils are moderately to poorly preserved throughout the succession. There appears to be a complete sequence of nannofossil zones from Zone NN6 (middle Miocene) through NP22 (lower Oligocene), except for Zone NN3, which could not be resolved. Planktonic foraminifers are present throughout the succession ranging from Zones N12 through O1. They are moderately well preserved in the Miocene and less well preserved in the Oligocene.

Benthic foraminifers are present throughout the section, although abundances are overall quite low. The preservation of tests is moderate in the upper part of Site U1336 (Samples 320-U1336A-1H-CC through 19H-CC, 8.22–170.63 m CSF, and 320-U1336B-1H-CC through 20H-CC, 1.68–174.01 m CSF) but deteriorates below this level. The Oligocene to middle Miocene benthic foraminifer assemblage is relatively diverse and indicates oligotrophic lower bathyal to abyssal paleodepths.

The Oligocene/Miocene boundary is placed between the first occurrence of Paragloborotalia kugleri (23.0 Ma) and the extinction of Sphenolithus delphix (23.1 Ma). The former occurs between Samples 320-U1336A-16H-CC and 17H-2, 38–40 cm (142.96 m CSF) and Samples 320-U1336B-16H-1, 52–54 cm, and 17H-3, 80–82 cm (137.72 m CSF). The top of S. delphix is recognized between Samples 320-U1336A-17X-2, 90 cm, and 17X-4, 90 cm (145.9 m CSF), and between Samples 320-U1336B-16H-CC and 17H-1, 150 cm (137.56 m CSF).

The radiolarian stratigraphy at Site U1336 spans the interval from just above the Zone RN6/RN5 boundary (middle Miocene) to the upper part of Zone RP22 (upper Oligocene) at ~170 m CSF. Below this level the sediments are barren of radiolarians. Above this level the assemblages tend to have good to moderate preservation with intermittent intervals of good preservation in Zones RN3 and RN4 (lower to middle Miocene). The downsection decrease in preservation and ultimate disappearance of the radiolarians below Core 320-U1336A-19H appears to be associated with dissolution and reprecipitation of the biogenic silica as intergranular cement and as chert.

Diatom stratigraphy in Hole U1336B spans the interval from just above the Cestodiscus peplum zone (middle Miocene) in Core 320-U1336B-1H to the lowermost part of the Crucidenticula nicobarica zone (upper lower Miocene) in Core 320-U1336B-7H. Below Sample 320-U1336B-7H-CC, the sediments are barren of diatoms. Above this level the valves tend to be mostly poorly preserved. Sample 320-U1336B-1H-CC contains the highest diversity with Cestodiscus pulchellus as dominant component, accompanied by Synedra jouseana and Thalassiosira yabei. Fragments of the large centric diatom Ethmodiscus are present in the upper part of Hole U1336B.

Paleomagnetic measurements were conducted on archive-half sections of 21 APC cores from Hole U1336A and 20 APC cores from Hole U1336B. Measurements of natural remanent magnetization (NRM) above ~80 m CSF in Holes U1336A and U1336B indicate moderate magnetization intensities (~1 × 10^–3 A/m) with a patchy but generally weak viscous remanent magnetization (VRM) or isothermal remanent magnetization (IRM) drilling overprint. Polarity reversal sequences are clearly recognized (Fig. F74). Demagnetization data from discrete samples above ~80 m CSF indicate that the characteristic remanent magnetization of the sediments is identified at the 10–20 mT demagnetization steps. The reversals pattern can be correlated with the GPTS from the base of Chrons C5r to C6n (~12 to 19 Ma).

Below ~80 m CSF, a zone of diagenetic alteration involving dissolution of remanence carriers reduces remanence intensities after AF demagnetization of 20 mT to values close to magnetometer noise level in the shipboard environment (~1 × 10^–5 A/m). In this zone, sediment magnetizations have been partly or entirely overprinted during the coring process and remanence inclinations are sometimes steep after AF demagnetization at peak fields of 20 mT. At ~130–140 m CSF (Cores 320-U1336A-15H through 16H and 320-U1336B-15H) and below ~160 m CSF (Cores 320-U1336A-19H through 21H and 320-U1336B-18H through 20H), polarity reversals are apparently present but the inclinations are steep (as much as 80°), indicating that the drilling overprint has not been effectively removed during shipboard demagnetization.

A complete physical property program was conducted on whole cores, split cores, and discrete samples comprising WRMSL measurements of magnetic susceptibility, bulk density, and P-wave velocity; NGR; and measurements of color reflectance, followed by discrete measurements of moisture and density properties, sound velocities, and thermal conductivity. Physical properties measurements on whole-round sections and samples from split cores reflect the differences among lithologies drilled at Site U1336 (Fig. F74). Nannofossil ooze with varying amounts of clay, radiolarians, and diatoms makes up lithologic Unit I and is characterized by high-amplitude and high-frequency variations in bulk density, magnetic susceptibility, NGR, and color reflectance. Magnetic susceptibility is highest in Unit I, with values ranging from 5 × 10^–5 to 30 × 10^–5 SI. NGR is also high in this unit, with values to 56 cps near the seafloor. Wet bulk densities are lowest in Unit I, with values ranging from 1.4 to 1.7 g/cm³. Porosity is highest in this interval, ranging from 65% to 80%. The grain density of most of the sediments of Unit I, as well as Units II and III, ranges from 2.6 to 2.9 g/cm³, reflecting the dominance of carbonate constituents at Site U1336. The sediment velocity in Unit I is low, averaging 1500 m/s. The color reflectance of Unit I is marked by luminance (L*) values that are slightly lower and more variable than values determined for sediments in Units II and III.

Below Unit I, a more uniform increase in wet bulk density and decrease in porosity in Units II and III reflects the increasing compaction of the sediments. A slight step increase in wet bulk density marks the transition between Units II and III. In Unit III wet bulk density and porosity average 1.9 g/cm³ and 51%, respectively. Magnetic susceptibility and NGR are low and nearly uniform in Units II and III. Magnetic susceptibility is typically below 5 × 10^–5 SI, and NGR is ~2 cps. Lower clay abundance in Unit II is marked by an increase in L* at the boundary between Units I and II. At 92 m CSF, within Unit II, sharp decreases in the a* and b* reflectance parameters mark the change in sediment color from pale yellow to greenish gray. One of the most pronounced changes in physical properties at Site U1336 is the sharp increase in velocity that accompanies the change from nannofossil ooze to nannofossil chalk at the boundary between Units II and III. The velocity at the base of Unit II is ~1700 m/s. Below 190 m CSF, in Unit III, the rate at which velocity increases with depth increases, ultimately reaching ~2200 m/s at 290 m CSF, near the base of Hole U1336A.

STMSL data were collected at 5 cm intervals from Hole U1336B and compared to the WRMSL data obtained at 2.5 cm resolution from Hole U1336A during Expedition 320. Features in the magnetic susceptibility and gamma ray attenuation density are well aligned between Holes U1336A and U1336B to a depth of ~94 m CCSF-A. Below 94 m CCSF-A, the magnetic susceptibility signal drops to very low values but the density data are good enough to sustain a correlation to interval 320-U1334B-14H-4, 122 cm. At this point (138.50 m CCSF-A) sediments recovered in both holes are disturbed.

Paleomagnetic reversals were used to calculate the average LSR for the upper 74 m of the section at Site U1336 on the corrected core composite depth below seafloor (CCSF-B; see "Corrected core composite depth scale" in the "Methods" chapter) depth scale. Below 74 m CSF only biostratigraphic datums were used to calculate the average LSR. The LSR at Site U1336 decreases from 15 m/m.y. in the upper Oligocene to 12 m/m.y. in the lower Miocene and stays relatively constant at 9 m/m.y. in the remainder of the section.

Standard geochemical analyses of pore water and organic and inorganic sediment properties were undertaken on Site U1336 samples. Alkalinity is relatively constant at values >2.5 mM in the upper 110 m CSF, with a pronounced decline to 1 mM by 170 m CSF. Sulfate concentrations decrease with depth to values as low as 22 mM. Dissolved manganese has a broad peak in the depth range from ~25 to 120 m CSF, and dissolved iron appears then peaks below 100 m CSF. The increase of dissolved iron occurs where Mn decreases downhole. Concentrations of dissolved silicate increase with depth from <400 to 800 µM.

Highlights

Miocene sedimentary section and cyclic sedimentation

One of the highlights from Site U1336 is the recovery of a thick Miocene carbonate section from the central equatorial Pacific, one of the high-priority objectives of the PEAT program. We recovered the complete early Miocene sequence (7.1 m.y. duration) in a ~110 m thick section, with a sedimentation rate of 12 m/m.y. and the middle Miocene sequence (4.4 m.y. duration) in a ~45 m thick interval with a sedimentation rate of ~21 m/m.y. These high sedimentation rates will facilitate the study of paleoceanographic processes at unprecedented resolution for the equatorial Pacific.

The obvious variations of both color and biogenic composition within nannofossil oozes represent cyclic fluctuations of CCD and upwelling intensity during the middle Miocene through early Miocene. The variable lithology also results in the variations of many petrophysical signals of physical properties including L*, b*, magnetic susceptibility, NGR, and GRA bulk density.

Oligocene–Miocene transition and depth transects

Site U1336 was planned as part of a latitudinal transect for early Miocene age slices and the PEAT Oligocene–Miocene depth transect compound in conjunction with Sites U1335 and U1337. The Miocene sequence at these sites includes the critical intervals of the Mi-1 glaciation and middle Miocene ice sheet expansion (Holbourn et al., 2005; Zachos et al., 2001b; Pälike et al., 2006b). The dominant lithologies of nannofossil ooze and chalk at Sites U1336 and U1335, with good preservation of calcareous microfossils, will allow us to achieve the prime objective for this site.

The Oligocene–Miocene transition in Hole U1336A occurs in homogeneous nannofossil ooze within the alternations of white and light greenish gray ooze. The same alternating sequence is observed above the Oligocene–Miocene transition at Site U1334. Biostratigraphy reveals that the Oligocene/Miocene boundary exists between 142.96 and 145.9 m CSF at Site U1336; this will allow the high-resolution study of this critical interval.

Geochemical front

Site U1336 recovered an interval of greenish gray carbonates that exhibit a distinct peak in dissolved Fe concentrations in pore water with similar characteristics as geochemical alteration fronts at Sites U1334 and U1335. At Site U1336, this zone is ~200 m thick. The remanent magnetization intensity is very weak in most parts of this section (80–180 m CSF). High dissolved Fe and Mn concentrations in pore water are caused by changes in the oxidation state of the sediments. The oxidation-reduction reactions are likely fueled by variable availability of organic carbon in the sediments. This site may provide the opportunity to study organic matter degradation.

Site U1336 migrated from south to north through the equatorial belt of high productivity. Based on paleolatitude reconstructions these geochemical alteration fronts can be mapped to similar equatorial positions between Sites U1334 and U1335, roughly between the Equator and ~4°N.

Chert formation in the early Oligocene

The sequence at Site U1336 includes barren intervals of radiolarian fossils and many thin intercalated chert layers and fragments. The radiolarians decrease in preservation downsection and disappear below Core 320-U1336A-19H. Instead, the sediments contain several chert fragments. Some inferred chert layers occur at ~120–140 m CSF and blocked APC penetration. Below ~190 m CSF, various colored chert layers and fragments occurred within the cores. The chert frequently contains foraminifer tests, reflecting diagenetic process of dissolution and reprecipitation of the biogenic silica.

The dissolution of biogenic silica is the source of porcellanite and chert and, on crust younger than 65 Ma, almost all cherts in the Pacific Ocean lie <150 m above basement. Although we did not recover basement rocks at this site, the sediments became hard, lithified limestones and the drilled section is probably close to basement. The dissolution of silica in the basal sedimentary section is likely associated with the circulation of warm hydrothermal waters in the upper oceanic crust that extend into the lower sediments where they are cut by fractures and faults (Moore, 2008a, 2008b). This site will provide information on chert formation in the equatorial Pacific regions.

Site U1337

The latest Oligocene through the middle Miocene appears to have been a time of relative warmth comparable to the latest Eocene. However, variability in the isotopic record of the early to middle Miocene is larger than that of the Eocene and may indicate more variability in climate and global ice volume. Site U1337 (proposed Site PEAT-7C; 3°50.009′N, 123°12.352′W; 4463 mbsl) (Fig. F75; see Table T1 in the "Site U1337" chapter) was targeted to collect an early middle Miocene segment of the PEAT equatorial megasplice on ~24 Ma crust between the Galapagos and Clipperton Fracture Zones, ~390 km southeast of Site U1335. In conjunction with Sites U1335 and U1336, it was also designed to provide a latitudinal transect for early Miocene age slices. The recovered sediment column at Site U1337 represents a nearly complete and continuous Neogene sedimentary section.

Operations

Four holes were cored at Site U1337. In Hole U1337A, APC cores were taken from the seafloor to 195.5 m DSF (Cores 321-U1337A-1H through 21H). Nonmagnetic core barrels were used for all APC cores except for Core 321-U1337A-21H. FlexIt core orientation was conducted for all cores except Core 321-U1337A-1H. In addition, five successful APCT-3 temperature measurements were taken with Cores 321-U1337A-5H, 7H, 9H, 11H, and 13H. XCB coring continued with Cores 321-U1337A-22X through 48X. The sediment/basement contact was recovered at the base of Core 321-U1337A-48X. Three logging strings (triple combo, FMS-sonic, and Versatile Seismic Imager [VSI]) were deployed in Hole U1337A.

In Hole U1337B, APC cores were taken from the seafloor to 245.2 m DSF (Cores 321-U1337B-1H through 27H). Nonmagnetic core barrels were used through Core 321-U1337B-20H. The FlexIt core orientation tool was deployed successfully for all but two APC cores (321-U1337B-17H and 18H). FlexIt and steel core barrels were used through Core 321-U1337B-27H. APCT-3 measurements were obtained with Cores 321-U1337B-15H, 17H, and 19H. Coring continued with a single XCB core (321-U1337B-28X) to 251.9 m DSF; however, this barrel could not be recovered and Hole U1337B was abandoned prematurely.

Hole U1337C was cored to recover sections that were missing from Holes U1337A and U1337B. APC cores were taken from the seafloor to 11.4 m DSF (Cores 321-U1337C-1H through 2H) using nonmagnetic core barrels and the FlexIt core orientation tool. A wash barrel (Core 321-U1337C-3W) was then deployed, and the hole was washed to 169.4 m DSF. APC coring resumed at that depth and continued through Core 321-U1337C-9H to 221.3 m DSF and then switched to steel core barrels. Coring with the XCB system continued with Cores 321-U1337C-10X through 33X. Basement was recovered in Core 321-U1337C-33X.

Hole U1337D was planned to target the few remaining areas that had yet to be fully recovered and to duplicate recovery through those sections of the formation already recovered to provide additional sample material. The most troublesome material encountered in the previous holes was the large diatom mats located directly above and below a hard ~0.4 m thick porcellanite layer. In Hole U1337D, APC cores were taken from the seafloor to 237.7 m DSF (Cores 321-U1337D-1H to 26H). Nonmagnetic core barrels were used through Core 321-U1337D-20H. The first XCB core (321-U1337D-27X) was designed to only core through the hard ~0.4 m thick porcellanite layer. The APC was once again deployed and cored to 267.0 m DSF (Cores 321-U1337D-28H through 30H). At this point the XCB coring system was once again deployed for Cores 321-U1337D-31X through 49X to a total depth of 442.9 m DSF. The FlexIt core orientation tool was deployed successfully with all APC cores. The SET was deployed for the first time from the JOIDES Resolution after Core 321-U1337D-17X at 298.1 m DSF.

Lithostratigraphy

At Site U1337, latest Oligocene seafloor basalt is overlain by ~450 m of nannofossil and biosiliceous oozes and nannofossil chalks that are divided into four lithologic units (Fig. F76). The Pleistocene through uppermost Miocene sediments of lithologic Unit I are characterized by multicolored (various hues of white, brown, green, and gray) nannofossil oozes, diatom oozes, and radiolarian oozes that alternate on meter scales with a general downsection increase in siliceous microfossils relative to nannofossils. The uppermost Miocene to middle Miocene lithologic Unit II is dominated by meter-scale interbeds of greenish gray biosiliceous sediments with white to light greenish gray nannofossil ooze. Within the unit are numerous diatom mat deposits. Meter-scale color alternations in Units I and II are associated with variations in lithology and physical properties. However, similar to the common millimeter- and centimeter-scale color banding that do not mark compositional changes, they are likely associated with sediment redox conditions. White, pale yellow, and pale green nannofossil oozes and chalks dominate the sediments of middle Miocene to latest Oligocene age, although diatoms and radiolarians remain present in low abundances. Seafloor basalt (lithologic Unit IV) was recovered at the base of the sedimentary section, dated as latest Oligocene.

Biostratigraphy

All major microfossil groups occur in the sediments recovered at Site U1337. Planktonic foraminifers at Site U1337 are rare to abundant with poor to good preservation throughout most of the succession but are absent or extremely rare in some intervals of the upper Miocene and lower Miocene. Biozones PT1b to O6 are recognized, with the exception of Zones PL4, M12, and M3 (Fig. F76). Calcareous nannofossils at Site U1337 are moderately to poorly preserved and some samples with high silica content are barren. Nannofossil Zones NN1 to NN21 are present, indicating an apparently complete sequence. The radiolarian stratigraphy at Site U1337 spans the interval from the uppermost part of Zones RN16–RN17 (upper Pleistocene) to RN1 (lower Miocene). The radiolarian assemblages of Pleistocene to upper Miocene age tend to have good preservation, whereas middle to lower Miocene assemblages show moderate preservation. In the lowermost part of the section, above the basement, sediments are barren of radiolarians. The high-resolution diatom stratigraphy at Site U1337 spans the interval from the Fragilariopsis (Pseudoeunotia) doliolus Zone (upper Pleistocene) to the lowermost part of the Craspedodiscus elegans Zone (lower Miocene). The diatom assemblage is generally well to moderately preserved throughout the recovered section; however, in several intervals valve preservation becomes moderate to poor. The base of the sediment column is barren of diatoms. The nannofossil, foraminifer, radiolarian, and diatom datums and zonal schemes generally agree, though some discrepancies occur in the lowest part of the core. Benthic foraminifers occur continuously throughout the succession recovered in Hole U1337A and show good to moderate preservation. The overall assemblage composition indicates lower bathyal to abyssal paleodepths.

Stratigraphic correlation

Stratigraphic correlation provided a complete spliced record to ~220 m CCSF-A. Several gaps were encountered over the next 50 m CCSF-A. Comparison of GRA density records with well logging density data suggest that no more than 1 m of section was lost in any of the gaps. Correlation between the holes was broken again several times between 440 m CCSF-A and basement at 490 m CCSF-A. Growth factor for the correlation was 1.12. The linear sedimentation rate decreases from ~21 m/m.y. in the middle Miocene to 17 m/m.y. in the late Miocene.

Paleomagnetism

Paleomagnetic measurements were conducted on archive-half sections of 20 APC cores and 14 XCB cores from Hole U1337A, 27 APC cores from Hole U1337B, 8 APC cores from Hole U1337C, and 30 APC cores from Hole U1337D. The FlexIt core orientation tool was deployed in conjunction with all APC cores, and we conclude that the FlexIt orientation data are generally reliable. Measurements of NRM above ~93 m CSF indicate moderate magnetization intensities (on the order of 10^–3 A/m) with a patchy but generally weak VRM or IRM drilling overprint, and polarity reversal sequences from Chrons C1n to C3r (0 to ~6 Ma) are recognized. Below ~93 m CSF, remanent intensities after AF demagnetization of 20 mT are reduced to values close to magnetometer noise level in the shipboard environment (~2 × 10^–5 A/m). In this zone, sediment magnetizations have been partly overprinted during the coring process, and remanent inclinations are occasionally steep after AF demagnetization at a peak field of 20 mT. Nonetheless, polarity reversals are apparently recorded to ~200 m CSF and are provisionally correlated to the GPTS from Chrons C3An to C5n (~6–11 Ma) (Fig. F76). Magnetic polarity interpretation was impossible for APC cores taken with steel core barrels and XCB cores because of severe magnetic overprint during coring.

Physical properties

Physical property measurements comprised WRMSL measurements of magnetic susceptibility, bulk density, and P-wave velocity; NGR; and measurements of color reflectance, followed by discrete measurements of moisture and density properties, sound velocities, and thermal conductivity. Physical property measurements on whole-round sections and samples from split cores display a strong lithology-dependent variation at Site U1337 (Fig. F76). Variations in the abundances of nannofossils, radiolarians, diatoms, and clay in lithologic Unit I account for high-amplitude, high-frequency variations of all physical properties. Intervals enriched in biogenic silica and clay generally display lower grain density and bulk density and higher porosity, magnetic susceptibility, and NGR. Velocity is generally directly related to bulk density; however, it is commonly higher in low-density siliceous-rich sediments than it is in more calcareous intervals. Wet bulk density is low in Unit I, ranging from 1.12 to 1.46 g/cm³. Porosity is as high as 92% in this unit. Velocity also is low, averaging 1525 m/s. The natural gamma record, as at previous sites, is marked by an anomalously high near-surface peak (~65 cps). Magnetic susceptibility varies between 4 × 10^–5 and 18 × 10^–5 SI. The color of Unit I is characterized by the lowest L* and high and variable a* and b* values. Lithologic Unit II is characterized by a continued high variability in grain density. Together, the grain density in Units I and II averages 2.51 g/cm³ and ranges from 2.17 to 2.85 g/cm³. All other physical properties display less variability in Unit II than in Unit I, reflecting a less variable lithology. Wet bulk density increases and porosity decreases with depth in Unit II; however, in Units II and III these trends are interrupted by low-density high-porosity diatom- and radiolarian-rich intervals. Unit II is slightly lighter colored (lower L*) and distinctly more blue (lower a*) and green (lower b*) than Unit I. Unit III is characterized by more uniform physical properties that accompany the high and uniform carbonate composition of the unit. The nannofossil oozes and chalks of this unit are characterized by a uniform grain density that averages 2.67 g/cm³. The bulk density and porosity trends of Unit II continue in Unit III. The transition from ooze to chalk is marked by a change in gradient of these properties to a more rapid decrease in wet bulk density and an increase in porosity with depth. Wet bulk density and porosity at the base of the sediment section are 1.95 g/cm³ and 47%, respectively. The increase in velocity with depth also changes to a higher gradient in Unit III, with values increasing from 1510 m/s at ~340 m CSF to ~1800 m/s near the base of the hole. Magnetic susceptibility and NGR values remain low in Unit III but do vary in response to small changes in lithology. The sharp color change from greenish gray to pale yellow at ~410 m CSF is marked by a sharp increase in a* and b*. The change in color to pale brown chalk immediately above basement is marked by an increase in both a* and b* and a decrease in L*.

Downhole logging

Three downhole logging tool strings were deployed in Hole U1337A. Two tool strings took measurements of NGR radioactivity, bulk density, electrical resistivity, elastic wave velocity, and borehole resistivity images in the 77–442 m WSF depth interval. The third tool string measured seismic waveforms in a VSP experiment in the 214–439 m WSF depth interval. Measurement depths were adjusted to match across different logging runs, obtaining a wireline log matched depth below seafloor (WMSF) depth scale. The downhole log measurements were used to define three logging units. Unit 1 (77–212 m WMSF) and Unit 2 (212–339 m WMSF) have average densities of ~1.3 and ~1.6 g/cm³, respectively, that do not show any trend with depth, whereas Unit III (339–442 m WMSF) density increases with depth, reaching 1.85 g/cm³ at the base of the hole (Fig. F77). Resistivity and P-wave velocity follow a pattern similar to that of density, suggesting that the major control on these physical properties are variations in sediment porosity. NGR measurements are low throughout the logged interval (~5 gAPI), except for two pronounced peaks caused by uranium, one at the seafloor and the other at 240 m WMSF. The gamma ray peak at 240 m WMSF corresponds to the ~40 cm thick pocellanite layer that has only been recovered as rubble in the cores but can be clearly identified in the downhole logs and borehole images as an interval of high density and resistivity. VSP logging measured arrival time of the seismic pulse from the sea surface at 16 stations. Together with the traveltime to the seafloor, VSP measurements are the basis for a traveltime-depth conversion that allows seismic reflectors to be correlated to stratigraphic events. Downhole temperature measurements and thermal conductivities of core samples were combined to estimate a geothermal gradient of 32.4°C/km and a heat flow of 28.4 mW/m² at Site U1337.

Geochemistry

A total of 85 interstitial water samples were collected from Hole U1337A, 49 using the whole-round squeezing approach across the entire hole and 36 in the upper 100 m by Rhizon sampling. Alkalinity increases slightly downhole from ~2.7 mM in the upper 100 m to values scattered around 3.8 mM below 300 m CSF. Sulfate concentrations vary between 26 and 29 mM, with slightly decreasing values with depth. A dissolved manganese peak of ~150 µM at 13 m CSF is captured by the high-resolution interstitial water sampling. Dissolved iron is sporadically detectable in the upper 200 m and then increases to a peak of ~5 µM between 275 and 300 m CSF before becoming undetectable again below 400 m CSF. These variations in manganese and iron reflect changes in redox chemistry that also manifest as changes in sediment color. Calcium carbonate and inorganic carbon concentrations were determined on 283 and 28 sediment samples from Holes U1337A and U1337B, respectively. Calcium carbonate contents vary greatly in the upper two lithologic units, ranging from 30 to 90 wt% (Fig. F76). In lithologic Unit III calcium carbonate contents are generally high, scattered around 80 wt%, but a distinctive decrease is observed between 350 and 400 m CCSF-A. In the upper 235 m CCSF-A, TOC content ranges between 0.10 and 0.34 wt% except for the high value of 0.72 wt% in the uppermost sample. TOC content increases at 44.00 m CCSF-A and in the interval from 87.28 to 108.59 m CCSF-A. Below 235 m CCSF-A, TOC values are generally <0.10 wt%.

Shipboard geochemical analyses of interstitial water and bulk sediment samples reflect large variations in sediment composition resulting from shifts in carbonate versus opal production. The large-scale redox state and diagenetic processes of the sediment column are related to overall changes in sediment composition. Interstitial water chemistry is also influenced by the porcellanite layer forming a barrier to diffusion at ~240 m CSF and by seawater circulation in the basement. The basement itself appears to exert little influence on the geochemistry of sediments and interstitial waters.

Highlights

Diatom mat deposition

Lithologic Unit II at Site U1337 is mostly composed of biosiliceous lithologies, notably diatoms. The abundance of diatoms in the middle and upper Miocene section at Site U1337 is much higher than encountered in any interval at Sites U1331–U1336. Several decimeter- to meter-scale intervals of diatom ooze are laminated, and smear slide analyses indicate that the diatom assemblage is composed primarily of pennate taxa, with abundant "needlelike" Thalassiothrix spp., indicating diatom mat deposition. The lowermost laminated diatom mat is in the upper portion of Unit III at ~15 Ma. Much thicker intervals are present in Unit II at roughly 10 Ma and shorter intervals at ~4.5 Ma. Ages of laminated diatom mats at this site are similar to those found at Leg 138 sites farther to the east (Mayer, Pisias, Janecek, et al., 1992), which have been interpreted to reflect regional bursts of silica export in the eastern equatorial Pacific (Kemp and Baldauf, 1993). No laminated diatom oozes were recorded during Expedition 320 at drill sites farther to the northwest.

Oligocene–Miocene transition

The Oligocene/Miocene boundary was recovered in Holes U1337A, U1337C, and U1337D. In Hole U1337A, the Oligocene/Miocene boundary is estimated to fall between Samples 321-U1337A-48X-2, 85–87 cm, and 48X-3, 55 cm (445.56–446.75 m CSF; 490.92–492.11 m CCSF-A). It occurs in white (2.5Y 8/1) nannofossil chalk with foraminifers, interbedded and heavily mottled with pale yellow (2.5Y 7/4) to very pale brown (10YR 7/4) nannofossil chalk. Abundant millimeter-scale dendritic manganese oxide grains occur throughout this interval. The lower 15 cm of the core catcher of Core 321-U1337A-48X is basaltic basement. No prominent change in lithology, GRA bulk density, reflectance, or magnetic susceptibility is seen through the Oligocene–Miocene transition.

Neogene carbonate dissolution

The CCD of the Neogene is much more stable than that of the Eocene, but there are intervals of lower carbonate deposition at Site U1337 that probably represent significant changes of the Neogene CCD. In the early Miocene, a significant carbonate low reaches its minimum at ~17 Ma (340 m CSF in Hole U1337A), when the site was at a depth of ~3500 meters below sea level. This early Miocene interval marks a strong minimum at Site U1334 as well, on crust with a depth of ~4000 m at that time. Highly variable carbonate is also characteristic of the late/middle Miocene boundary interval, but the role of carbonate dissolution versus elevated deposition of biosilica needs to be determined.

Site U1338

Site U1338 (proposed Site PEAT-8D; 2°30.469′N, 117°58.178′W; 4200 mbsl) (Fig. F78; see Table T1 in the "Site U1338" chapter) was sited to collect a 3–18 Ma segment of the PEAT equatorial megasplice and is located on ~18 Ma crust just north of the Galapagos Fracture Zone, 324 nmi (600 km) southeast of Site U1337 (Fig. F78). A seamount (3.7 km water depth) with surrounding moat is found ~25 km north-northwest of Site U1338 at the downslope end of the survey area. Originally a site was chosen ~10 km from the seamount (proposed Site PEAT-8C). However, the alternate proposed site was selected and drilled uphill and further away from the seamount to avoid possible turbidites, as were found near seamounts during drilling of Expedition 320 Sites U1331 and U1335. The recovered sediment column at Site U1338 represents a nearly complete and continuous lower Miocene to Holocene sedimentary section.

Operations

Four holes were cored at Site U1338. From Hole U1338A, APC cores were taken from the seafloor to 221.2 m DSF (Cores 321-U1338A-1H through 24H) using nonmagnetic core barrels and the FlexIt core orientation tool installed. FlexIt and steel core barrels were used for Cores 321-U1338A-25H and 26H. In addition, five successful APCT-3 temperature measurements were taken with Cores 321-U1338A-5H, 7H, 9H, 11H, and 13H. XCB coring continued with Cores 321-U1338A-27X through 44X. A small piece of basement was recovered in the core catcher of Core 321-U1338A-44X.

From Hole U1338B, APC cores were taken from the seafloor to 188.1 m DSF (Cores 321-U1338B-1H through 20H) except for a short drilled interval of 2.5 m from 235.6 to 238.1 m DSF to adjust the core breaks. Nonmagnetic core barrels and the FlexIt core orientation tool were used through Core 321-U1338B-20H. FlexIt and steel core barrels continued through Core 321-U1338B-42H to 387.4 m DSF. Coring continued with three XCB cores (321-U1338B-43X through 45X) to 416.1 m DSF. Basement contact was recovered in Core 321-U1338B-45X. Three logging strings (triple combo, FMS-Sonic, and VSI) were deployed in Hole U1338B.

Hole U1338C was cored to recover sections that were missing from Holes U1338A and U1338B. APC cores were taken from the seafloor to 189.8 m DSF (Cores 321-U1338C-1H through 21H) using nonmagnetic core barrels and the FlexIt core orientation tool. FlexIt and steel core barrels were used through Core 321-U1338C-44H to 396.9 m DSF. Coring continued through Core 321-U1338C-47H to a total depth of 414.4 m DSF, which, at the time, set a new all time depth record for the APC.

Hole U1338D was primarily planned to recover a few "instructional" cores to be used during Expedition 323. Three APC cores were cut to 23.9 m DSF.

Lithostratigraphy

At Site U1338, ~415 m of nannofossil ooze and chalk with varying concentrations of diatoms and radiolarians overlie the seafloor basalt and are divided into three lithologic units (Fig. F79). Pleistocene through middle Pliocene sediments of Unit I are characterized by multicolored (various hues of white, brown, green, and gray) nannofossil ooze, diatom nannofossil ooze, and radiolarian nannofossil ooze that alternate on a decimeter to meter scale. Light green and light gray nannofossil ooze with occasional darker intervals with abundant siliceous microfossils, notably diatoms, comprise the upper Miocene to middle Pliocene Unit II. Decimeter-, meter- and tens of meters–scale color alternations in Units I and II are associated with variations in lithology and physical properties. Some of these color changes, as well as common millimeter- and centimeter-scale color banding, are not associated with compositional changes and likely reflect variations in sediment redox state. White, pale yellow, light greenish gray, and very pale brown nannofossil oozes and chalks dominate Unit III of the lower to upper Miocene, although slightly darker green and gray intervals with larger amounts of siliceous microfossils remain present. Seafloor basalt (Unit IV) was recovered at the base of the sedimentary section, overlain by lower Miocene sediments.

Biostratigraphy

All major microfossil groups have been found in the ~415 m thick succession of Holocene to lower Miocene sediment bulge recovered from Site U1338. Calcareous nannofossils at Site U1338 are in general moderately preserved, but there are some intervals in which the preservation is good or poor. Nannofossil Zones NN4 to NN21 are present, indicating an apparently complete sequence. Planktonic foraminifers vary from rare to abundant, with moderate to good preservation throughout most of the succession, but are absent or rare in a short interval in the upper Miocene. Planktonic foraminifer Zones PT1b (upper Pleistocene) to M2 (lower Miocene) are documented, with the exception of Zones PL4, M12, and M6. The radiolarian stratigraphy spans the interval from the uppermost part of Zones RN16–RN17 (upper Pleistocene) to the uppermost part of Zone RN3 (lower Miocene). Radiolarian assemblages show good to moderate preservation except in the lowermost portion (lower Miocene), which is barren of radiolarians. The high resolution diatom stratigraphy spans the interval from the Fragilariopsis (Pseudoeunotia) doliolus Zone (upper Pleistocene) to the lowermost part of the Craspedodiscus elegans Zone (lower Miocene). The diatom assemblage is generally well to moderately preserved throughout the recovered section; however, there are several intervals in which valve preservation becomes moderate to poor. The nannofossil, foraminifer, radiolarian, and diatom datums and zonal schemes generally agree, with some inconsistencies (Fig. F79). Benthic foraminifers occur continuously throughout the succession recovered in Hole U1338A and show generally good preservation. The overall assemblage composition indicates lower bathyal to abyssal paleodepths.

Stratigraphic correlation

Stratigraphic correlation provided a complete spliced record to a depth of ~260 m CCSF-A. Several gaps were seen between 280 and 360 m CCSF-A. Comparison of GRA density records with well logging density data suggests that no more than 1 m of section was lost in any of the gaps. Correlation between the holes became difficult again several times between 435 m CCSF-A and basement at 460 m CCSF-A. The growth factor for the correlation was 1.11. The linear sedimentation rate decreases from ~29 m/m.y. in the Miocene to 13 m/m.y. in the Pliocene–Pleistocene.

Paleomagnetism

Paleomagnetic measurements were conducted on archive-half sections of 26 APC cores from Hole U1338A, 42 APC cores from Hole U1338B, and 47 APC cores from Hole U1338C. The FlexIt core orientation tool was deployed in conjunction with all APC cores except for the deepest three cores of Hole U1338C, and we conclude that the FlexIt orientation data are generally reliable. NRM measurements indicate moderate magnetization intensities (on the order of 10^–3 A/m) for depth intervals 0–50, 280–225, and 295–395 m CSF. Polarity reversal sequences of these intervals are provisionally correlated to Chrons C1n to C2Ar (0 to ~4 Ma), Chrons C4An to C5n (~9–11 Ma), and Chrons C5r to C5Br (~12–16 Ma) of the GPTS, respectively (Fig. F79). Except for these intervals, remanent magnetic intensities after AF demagnetization of 20 mT are reduced to values close to magnetometer noise level in the shipboard environment (~2 × 10^–5 A/m). Magnetization directions are dispersed and not interpretable there.

Physical properties

Physical property measurements comprised WRMSL measurements of magnetic susceptibility, bulk density, and P-wave velocity; NGR; and measurements of color reflectance, followed by discrete measurements of moisture and density properties, sound velocities, and thermal conductivity. Physical property measurements on whole-round sections and samples from split cores display a variation strongly dependent on the relative abundance of biosiliceous and calcareous sediment components at Site U1338. As at Site U1337, intervals enriched in siliceous microfossils and clay generally display darker colors, lower grain density and bulk density, and higher porosity, magnetic susceptibility, and NGR. The variation of velocity is more complex in that it is dependent on both the wet bulk density and the sediment rigidity. These parameters vary independently with the variation in abundance of biosiliceous and calcareous components. The physical properties at Site U1338 also display cyclicity on multiple scales, a decimeter to meter scale and a scale with a spacing on the order of tens of meters.

Lithologic Unit I at Site U1338 is characterized by low wet bulk density that decreases from 1.4 g/cm³ near the seafloor to 1.2 g/cm³ at the base of the unit as a result of an increasing abundance of radiolarians and diatoms with depth. The grain density in Units I and II displays a greater variability than is found deeper at the site as a result of the greater variability in the abundance of biosiliceous and calcareous components. The average grain density for Units I and II is relatively low, at 2.59 g/cm³. The NGR signal at Site U1338 is characterized by a near-seafloor peak that is somewhat lower than those recorded at the other PEAT drill sites but extends deeper and is marked by a double peak. Spectral reflectance measurements show that Unit I is characterized by lower L* and higher a* and b* values in the upper 25 m of Unit I (Fig. F79). Below 25 m CSF, the sediment becomes lighter colored (L* increases) and more bluish green (a* and b* decrease).

Unit II is characterized by increasing wet bulk density with depth to ~175 m CSF. Below this depth, an increase in the abundance of siliceous microfossils produces a broad density minimum. Magnetic susceptibility and NGR signals are low in Unit II to the depth at which the biosiliceous material increases in abundance. The interval of the broad density minimum is characterized by higher magnetic susceptibility values that are roughly equal to those in the upper 25 m of Unit I. Unit II is lighter colored than Unit I (higher L*) and more blue (lower b*).

Unit III at Site U1338 is characterized by a higher and more uniform carbonate content and, as a result, more uniform physical properties. Wet bulk density increases from ~1.5 g/cm³ at the top of Unit III to 1.7 g/cm³ at the base of the unit. Grain density varies over a narrower range in Unit III than it does in Units I and II and displays an average (2.64 g/cm³) nearer to that of calcite. Velocity, which through much of Units I and II is close to the velocity of water, displays a regular increase in Unit III, from ~1620 m/s at the top to ~1820 m/s near the base of the unit. Velocity gradient increases near the base of Unit III accompanying the transition from nannofossil ooze to chalk. Magnetic susceptibility is low from the boundary between Units II and III, at ~245 m CSF, to 300 m CSF. Below 300 m CSF, susceptibility again increases to values comparable to those in the upper part of Unit I. NGR variability is lower in Unit III than in Unit II and remains uniformly low throughout the unit. Overall, Unit III is the lightest colored (highest L* values) unit at Site U1338. The transition from greenish gray to pale yellow is marked at ~385 m CSF by a shift to higher values of both a* and b*.

Downhole logging

Three downhole logging tool strings were deployed in Hole U1338B: a modified triple combo (that did not include a neutron porosity measurement), an FMS-sonic combination, and a VSI seismic tool with a Scintillation Gamma Ray (SGT-N) sonde. The modified triple combo and FMS-sonic tool strings took downhole measurements of natural gamma ray radioactivity, bulk density, electrical resistivity, elastic wave velocity, and borehole resistivity images in the 125–413 m WSF depth interval. The VSI seismic tool string measured seismic waveforms in a VSP experiment that covered the 189.5–414.5 m WSF depth interval. Measurement depths were adjusted to match across different logging runs, obtaining the WMSF depth scale.

Downhole log measurements were used to define three logging units: Unit I (139–244 m WMSF) and Unit II (244–380 m WMSF) have average densities of ~1.45 and ~1.6 g/cm³, respectively, that do not show any trend with depth, whereas in Unit III (from 380 m WMSF) density increases with depth, reaching 1.7 g/cm³ at the base of the hole (Fig. F80). Resistivity and P-wave velocity follow a pattern similar to that of density throughout the logged interval, suggesting that the major control on these physical properties are variations in sediment porosity. Both resistivity and density measurements show a small-scale peak at 280 m WMSF. This peak at 280 m WMSF is clearly visible in the borehole resistivity images as a high-resistivity layer 16 cm thick, and it corresponds to a chert layer that has only been recovered as rubble in the cores. Natural gamma ray measurements are low throughout (~4 gAPI) but do show a pronounced high at the seafloor caused by a local increase in uranium concentration.

In the VSP experiment, the arrival time of a seismic pulse was measured from the sea surface at 14 stations. Together with the traveltime to the seafloor, the VSP measurements are the basis for a traveltime-depth conversion that allows seismic reflectors to be correlated to stratigraphic events. Downhole temperature measurements and thermal conductivities of core samples were combined to estimate a geothermal gradient of 34.4°C/km and a heat flow of 33.6 mW/m² at Site U1338.

Geochemistry

A standard shipboard suite of geochemical analyses of pore water and organic and inorganic sediment properties was undertaken on samples from Site U1338. Alkalinity increases slightly downhole from ~2.7 mM at the sediment/water interface to peak slightly above 4 mM at 140 m CSF. A large dissolved manganese peak of 150 mM at 10 m CSF is captured by the high-resolution interstitial water sampling and is remarkably similar to that observed at Site U1337. These peaks are >10 times larger than the highest dissolved manganese concentrations encountered during Expedition 320. Lithium concentrations decrease from ~26 µM at the surface to a minimum of ~3 µM at ~250 m CSF before increasing sharply with depth to seawater values at the base of the section. The interstitial water strontium profile is a mirror image to that of lithium except the decrease from the peak of 400 µM at 200 m CSF is punctuated by a sharp drop of >100 µM between ~260 and 290 m CSF. The lithium and strontium profiles indicate seawater circulation in the basement as their values tend toward seawater values near the basement.

Calcium carbonate contents range between 26 and 88 wt% with substantial variability in the upper 273.31 m CCSF-A, corresponding to the alternation between calcite and opal production in the upper two lithologic units. Below 273.31 m CCSF-A (lithologic Unit III), calcium carbonate contents become relatively high and stable between 66 and 91 wt% compared with the upper part of the stratigraphic column (Fig. F79). In the upper ~230 m CCSF-A, TOC content is generally high and variable ranging between 0.09 and 0.46 wt%, whereas below ~230 m CCSF-A, TOC content is <0.09 wt%. Downhole TOC variability is most likely related to lithologic changes, with higher TOC being found in the more biosiliceous intervals.

Interstitial water and bulk sediment geochemistry reflect large variations in sediment composition resulting from shifts between carbonate and opal dominance. The large-scale redox state and diagenetic processes of the sediment column are related to overall changes in sediment composition. Interstitial water chemistry points to seawater circulation in the basement, although the basement itself appears to exert little influence on the geochemistry of the sediments and interstitial waters.

Highlights

Color changes, lithology, and redox state

Smear slide analyses and visual core descriptions show that many of the decimeter-, meter-, and tens of meters–scale color variations in lithologic Units I and II to some extent relate to changes in lithology (e.g., Fig. F79). We suspect, however, that some of these color variations, notably the transitions between pale green and pale yellow lithologies, are controlled by sediment redox state, similar to those recorded at Sites U1331–U1337 and earlier work in the equatorial Pacific Ocean (e.g., Lyle, 1983).

Magnetic susceptibility is moderately low in the light gray and light brown intervals in Unit I (Fig. F79). A significant decrease in magnetic susceptibility in Unit II suggests dissolution of magnetite resulting from intensified microbial Fe reduction. In the lower part of Unit III, a sharp downcore transition from green to yellow is not associated with any other lithologic change, does not occur at the same stratigraphic level between holes, and thus should not be considered as an equivalent time horizon. Pore water Fe concentrations reach 6 to 7 µM in the green interval, and Fe is absent below the transition to yellow and brown. Although some of this signal may be affected by seawater contamination during XCB drilling, all available information suggests that the lowermost color change represents a redox front.

Occurrence of diatom-rich layers

Lithologic Unit II at Site U1338 is mainly composed of nannofossil ooze with relatively high abundances of biosiliceous components, notably diatoms (Fig. F79). The relative abundance of diatoms is lower than that at Site U1337, and the record lacks laminated diatom ooze intervals (diatom mats) such as those observed at Site U1337. However, centimeter to sometimes 1–2 m thick diatom nannofossil ooze layers containing abundant specimens of Thalassiothrix spp. are occasionally interbedded with nannofossil ooze (e.g., ~126.2–127.1 and ~231.8–234.3 m CSF in Hole U1338A and ~127.3–128.0 and ~233.8–234.8 m CSF in Hole U1338C). Units II and III also contain significant amounts of pyrite, particularly in diatom-rich intervals in Unit II (e.g., Cores 321-U1338B-14H, 19H through 21H, 26H, 28H, 29H, and 32H through 41H). In addition, the middle part of Unit III contains thin intervals of abundant pyrite-filled siliceous microfossils (e.g., intervals 321-U1338B-33H-4, 58–66 cm, and 35H-5, 76–82 cm). These diatom-rich layers, pyrite nodule occurrences, and pyrite-rich siliceous microfossil layers in Units II and III are associated with high TOC content, suggesting a relation between the abundance of diatoms in the sediments, sediment redox state, and the export or preservation of organic carbon.

Top of page   |   Previous   |   Next