Previous   |   Next

doi:10.2204/iodp.proc.320321.110.2010

Biostratigraphy

At Site U1338, a ~415 m thick succession of Holocene to lower Miocene sediments was recovered. Calcareous nannofossils at Site U1338 are in general moderately preserved, but in some intervals the preservation is good or poor. Nannofossil Zones NN21 to NN4 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 the interval from Sample 321-U1338A-22H-5, 41–43 cm, to 24H-CC. Planktonic foraminifer Subzone PT1b (upper Pleistocene) to Zone M2 (lower Miocene) are documented, with the exception of Zones PL4, M12, and M6. Radiolarian stratigraphy spans the interval from the uppermost part of Zone 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. 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 minor inconsistencies. The integrated calcareous and siliceous microfossil biozonation is shown in Figure F13. An age-depth plot including biostratigraphic and paleomagnetic datums is shown in Figure F14. 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. Marked variations in downcore abundance and species distribution may reflect major changes in global climate linked to fluctuations in ice volume and reorganization of Pacific Ocean circulation during the Neogene.

Calcareous nannofossils

Calcareous nannofossil biostratigraphy indicates that a continuous sediment sequence was recovered at Site U1338. Zones NN20–NN21 through lower NN4 are recognized in all three holes cored (Table T4). Calcareous nannofossil biostratigraphic markers are often determined to ±0.75 m in all three holes. The oldest biostratigraphic marker observed is the first occurrence of common Sphenolithus heteromorphus at 403.2 ± 0.4 m CSF in Hole U1338A, ~5 m above the sediment/basalt contact. Zones NN15 through NN13 are not differentiated. The Zone NN8/NN9 boundary is readily recognized by the first occurrence of Discoaster hamatus, but the base of Zone NN8 is poorly constrained because of discontinuous and exceptionally rare occurrences of members of the genus Catinaster. Similarly, the base of Zone NN5 is poorly constrained because of discontinuous and rare occurrences of Helicosphaera ampliaperta.

Calcareous nannofossils generally make up between 50% and 90% of the biogenic particles in the smear slides investigated. Exceptions include: (1) the uppermost 12–13 m of the sediment sequence where nannofossils are dominant (>90% of the fossils) and (2) a biosilica-rich interval around the middle/late Miocene boundary where nannofossils are less abundant and make up ~10%–50% of the biogenic particles in several samples. Preservation follows abundances, with good preservation in the uppermost 12–13 m where nannofossils dominate, moderate preservation in the majority of samples investigated, variable preservation in the interval with biosilica-rich sediments, and generally poor preservation in the chalks of Zone NN4. Members of the genus Discoaster vary markedly in terms of composition and abundance. Their preservation is often good in the more dissolved samples. Many of the morphotypes observed have been grouped under the Discoaster variabilis concept, although it is likely that several species are included in this group. Discoasters in upper Miocene sediments exhibit large mophologic variability and taxonomic diversity, which is only partially resolved at the species level here. Middle and lower Miocene discoasters are generally made indistinct by overgrowth of secondary calcite. Abundances of all calcareous nannofossil taxa and preservation of samples are summarized in range chart format from Hole U1338A, including a few samples from Hole U1338B, in Table T5.

Pleistocene sediments are recognized by a series of extinctions, from top to bottom: Pseudoemiliania lacunosa, Reticulofenestra asanoi, Gephyrocapsa spp. >5.5 µm, and Calcidiscus macintyrei. The Pleistocene assemblages are partly characterized by a complete dominance of small gephyrocapsids (<3 µm), for example in Samples 321-U1338A-2H-2, 24 cm, and 2H-5, 120 cm, outnumbering other taxa by at least an order of magnitude. The undifferentiated Zones NN20/NN21 are only recognized in Sample 321-U1338A-1H-CC.

Pliocene bioevents include the extinctions of Discoaster brouweri (top Zone NN18), Discoaster pentaradiatus (top Zone NN17), Discoaster surculus (top Zone NN16), Sphenolithus abies, and Reticulofenestra pseudoumbilicus (top Zone NN15). D. pentaradiatus and D. surculus disappear together in the 2 m sample distance used in the pertinent interval, thereby eliminating Zone NN17. The latter species has a more distinct last occurrence. D. pentaradiatus is continuously present only in a ~40 m long interval across the Miocene/Pliocene boundary and shows discontinuous occurrences in middle Pliocene and lower upper Miocene sediments. Some lower Pliocene samples (e.g., Sample 321-U1338A-7H-CC) contain quite large (>15 µm) specimens of D. pentaradiatus. The onset of abundant Discoaster triradiatus is observed in Section 321-U1338C-4H-1.

Sample 321-U1338A-7H-CC contains a nannolith that resembles transitional forms between the middle Miocene Discoaster micros and Catinaster spp. The lower Pliocene morphotypes in Core 321-U1338A-7H are likely indigenous, as neither D. micros nor Catinaster spp. occur in any corresponding abundance in lower upper Miocene sediments, nor is there evidence of lower upper Miocene nannofossil reworking. Discoaster altus is also present in these lower Pliocene sediments. A lack of R. pseudoumbilicus is observed in a few samples at the Miocene/Pliocene boundary (upper interval of Core 321-U1338A-9H, lower interval of Core 8H). A longer interval in upper Miocene sediments also shows a paracme episode of this species (Sections 321-U1338A-16H-4 through 20H-3). The underlying reason(s) causing these two paracme intervals is probably similar.

Zone NN11 is 75 m thick in the interval between the extinction of Discoaster quinqueramus in Sample 321-U1338A-10H-3, 70 cm, and the appearance of Discoaster berggrenii in Sample 321-U1338A-18H-3, 65 cm, suggesting an average late Miocene sedimentation rate of 27.5 m/m.y. Assemblages in Zone NN11 are moderately preserved, with abundant nannofossils. Bioevents observed within this zone include the top and basal occurrence of Nicklithus amplificus, the top of the paracme interval of R. pseudoumbilicus, and the base of Amaurolithus spp. Members of the genus Discoaster are generally well preserved and richly represented in these upper Miocene sediments. Several different morphotypes are grouped in a broadly applied D. variabilis concept, implying that the taxonomic diversity of late Miocene discoasters is not fully resolved in this report.

Samples from the base of Core 321-U1338A-19H and the top of Core 20H have moderate to good preservation and are dominated by Dictyococcites spp. <3 µm and small Sphenolithus specimens. These assemblages have few discoasters and reticulofenestrids, common calcidiscids, and no dictyococcitids or reticulofenestrids >5 µm. The onset of the paracme interval of R. pseudoumbilicus occurs over <1.5 m in lower Core 321-U1338A-20H. The silica content increases downhole from Cores 321-U1338A-20H to 25H, but in general nannofossils are abundant and moderately preserved. In this lower upper Miocene interval, Discoaster diversity and abundance increases with several species of five- and six-rayed forms present, whereas the placolith assemblage composition and diversity remains constant. Zone NN9 is placed between Samples 321-U1338A-22H-4, 80 cm, and 25H-1, 80 cm, defined by the total range of D. hamatus. The stratigraphically highest occurrence of Coccolithus miopelagicus is observed from just below the base of D. hamatus, similar to that observed at Site U1337 (see "Biostratigraphy" in the "Site U1337" chapter for discussion). A few specimens of Catinaster coalitus are observed in Core 321-U1338A-25H, indicating an age not older than Zone NN8 for this core. Extensive searches for Catinaster spp. in corresponding cores from Holes U1338B and U1338C reinforced the initial findings from Hole U1338A, which were that members of this genus occur only sporadically in the occasional sample. This absence is likely not caused by selective calcite dissolution, judging from the abundant occurrences of other calcareous nannofossils such as discoasters, various placoliths, sphenoliths, and umbilicosphaerids. Catinasters occur abundantly in sediments of coeval age at ODP Site 926 in the western Atlantic Ocean (Backman and Raffi, 1997) at a time when Atlantic and Pacific Ocean surface waters were connected by the Central American seaway (Keigwin, 1978). The similar positions of Site 926 (3°43′N) and Site U1338 (2°30′N) with respect to latitude suggests that temperatures in the photic zone were similar at the two near-equatorial sites. Some other factor was involved that resulted in the ecological exclusion of catinasters from the calcareous nannoplankton assemblages at Site U1338. A clue is provided by the absence of diatoms at Site 926 and the high abundance of this group at Site U1338, which may suggest that catinasters occupied the same ecological niche as the prolific diatoms and simply were outcompeted.

The preservation in Cores 321-U1338A-26H and 27X varies from poor to good depending upon the amount of silica present. An assemblage in lower Core 321-U1338A-27X is characterized by abundant to dominant small placolith species and few discoasters. Core 321-U1338A-28X had only 2% recovery, and the top of common Discoaster kugleri occurs in the missing interval. In Hole U1338B, the top of common D. kugleri occurs between Samples 321-U1338B-27H-2, 80 cm, and 28H-1, 80 cm, but this datum cannot be resolved further in this hole because of severe core disturbance (flow-in) from interval 321-U1338B-27H-2, 100 cm, to Sample 321-U1338B-27H-CC (see "Lithostratigraphy"). The best constrained determination of the top of D. kugleri is from Hole U1338C, where it is placed to within ±0.26 m. The first appearance of common D. kugleri (Zone NN6/NN7 boundary) occurs at the base of Core 321-U1338A-29X, which corresponds to Core 321-U1338B-29H. The poor recovery in Cores 321-U1338A-30X (1%) and 31X (5%) implies that datums correlated to upper Zone NN6 are lost in the unrecovered intervals. These datums are, however, determined in Hole U1338B. The extinctions of Cyclicargolithus floridanus and Coronocyclus nitescens, separated by only 80 k.y., both occur between Samples 321-U1338B-29H-2, 80 cm, and 29H-3, 80 cm. The top of Calcidiscus premacintyrei occurs ~3 m further downcore.

Sediment from Zone NN6 (Cores 321-U1338A-30X through 35X) contains abundant moderately preserved nannofossils, but pervasive overgrowth is present throughout, making discoasters difficult to identify. The top of common C. floridanus is placed at the base of Core 321-U1338A-34X. Several changes in the Triquetrorhabdulus lineage occur in Zone NN6 (Fig. F15). The species present in Zones NN4 and NN5 are Triquetrorhabdulus milowii and Orthorhabdus serratus, the latter of which has a sporadic presence. The evolutionary appearance of Triquetrorhabdulus rioi is observed at the base of Zone NN6 (Sample 321-U1338A-34X-CC). This taxon has a continuous presence throughout Zone NN6 and has a frequent to common abundance. Orthorhabdus serratus has an elevated abundance throughout the samples from Zone NN6, whereas T. milowii gradually decreases in its occurrence toward the top of Zone NN6. The first common occurrence of Triquetrorhabdulus rugosus occurs in Sample 321-U1338B-29H-4, 80 cm. Below this sample T. rugosus is rare in Zone NN6. Both T. rioi and T. rugosus have a notable first common occurrence, which was also observed by Raffi et al. (1995), potentially making these events useful biostratigraphic datums.

The stratigraphically highest occurrence of S. heteromorphus occurs in Sample 321-U1338A-35X-CC, marking the Zone NN5/NN6 boundary. Cores 321-U1338A-35X through 37X all had minimal recovery (1%–3%). The assemblages observed in the core catchers tend to be diverse and overgrown. The extinction of H. ampliaperta, which defines the Zone NN4/NN5 boundary, is difficult to recognize because the marker is rare, with a handful of specimens observed in only four samples (Table T5). A single etched specimen of H. ampliaperta, which may possibly be reworked, is observed in Sample 321U1338A-38X-CC. The top of common D. deflandrei and the base of Discoaster petaliformis, referred to as Discoaster signus in Bukry (1971), occur with a short overlap (<2.8 m). These two events are only 34 k.y. apart in the western equatorial Atlantic Ocean (Raffi et al., 2006). The base of Hole U1338A (Cores 321-U1338A-40X through 44X) contains carbonate-rich sediment with abundant nannofossils. The preservation in this interval is poor with extensive overgrowth. Despite the preservational state, nannofossil assemblages are diverse and contain multiple species of Helicosphaera and Scyphosphaera. The base of common S. heteromorphus is in Core 321-U1338A-44X, located 5 m above the sediment/basalt contact. At a sedimentation rate of 20 m/m.y., this makes the basement age 17.96 Ma (0.25 m.y. older than the base of common S. heteromorphus).

Radiolarians

High-resolution radiolarian biostratigraphy at Site U1338 was performed on core catchers and additional samples (mostly two per core) from Hole U1338A. Six samples from Cores 321-U1338B-27H through 30H were also used because the corresponding interval was not recovered in Hole U1338A.

The radiolarian stratigraphy at Site U1338 (Table T6) spans the interval from Zones RN16–RN17 (upper Pleistocene) to RN3 (lower Miocene) (Fig. F13). Below Sample 321-U1338A-44X-2, 44–46 cm (uppermost portion of Zone RN3), sediments are barren of radiolarians. Assemblages tend to show good to moderate preservation. Upper upper Miocene to lowermost Pleistocene sediments contain rare reworked radiolarian individuals from the early to middle Miocene. Abundances of all radiolarian taxa and preservation of samples are summarized in Tables T7 and T8.

The top of Hole U1338A corresponds to Zone RN16–RN17 (Pleistocene). The base of Zone RN16–RN17 is placed between Samples 321-U1338A-1H-CC and 2H-2, 67–69 cm, as indicated by the top of Stylatractus universus. The base of Collosphaera tuberosa, which indicates the boundary between Zones RN15 and RN14, also occurs in the same interval. The top of Anthocyrtidium angulae occurs between Samples 321-U1338A-2H-5, 110–112 cm, and 2H-CC, marking the base of Zone RN14.

The top of Theocyrtidium vetulum is between Samples 321-U1338A-2H-5, 110–112 cm, and 2H-CC. The last occurrence of Lamprocyrtis heteroporos and the first occurrences of Theocyrtidium tracherium and A. angulae are both distinguished between Samples 321-U1338A-3H-2, 67–69 cm, and 3H-5, 110–112 cm. The top of Pterocanium prismaticum occurs between Samples 321-U1338A-3H-5, 110–112 cm, and 3H-CC and defines the base of Zone RN13. The top of Didymocyrtis avita is recognized between Samples 321-U1338A-4H-4, 34–36 cm, and 4H-CC. The top of Stichocorys peregrina is observed between Samples 321-U1338A-4H-CC and 5H-2, 123–124 cm, marking the boundary between Zones RN12 and RN11. The base of L. heteroporos is distinguished between Samples 321-U1338A-5H-2, 123–124 cm, and 5H-4, 50–51 cm. The range of Lamprocyrtis neoheteroporus and top of Anthocyrtidium pliocenica are not determined because of their sporadic and discontinuous occurrence in Hole U1338A.

The top of Phormostichoartus fistula is recognized between Samples 321-U1338A-5H-4, 50–51 cm, and 5H-CC and defines the boundary between Subzones RN11b and RN11a. Both the top of Spongaster pentas and top of Phormostichoartus doliolum are observed between Samples 321-U1338A-6H-3, 131–133 cm, and 6H-5, 131–133 cm. The latter datum marks the top of Zone RN10. The base of Amphirhopalum ypsilon is observed between Samples 321-U1338A-6H-5, 131–133 cm, and 6H-CC. The top of Spongaster tetras tetras is between Samples 321-U1338A-6H-CC and 7H-2, 47–49 cm.

The top of Zone RN9 is indicated by the base of Didymocyrtis penultima, which occurs between Samples 321-U1338A-6H-CC and 7H-2, 47–49 cm. The base of Pterocanium prismaticum is recognized between Samples 321-U1338A-7H-2, 47–49 cm, and 7H-4, 52–54 cm. The top of Spongaster berminghami is observed between Samples 321-U1338A-8H-CC and 9H-3, 14–16 cm. The lowermost occurrences of Didymocyrtis avita and S. pentas and the top of Didymocyrtis antipenultima are distinguished between Samples 321-U1338A-10H-6, 118–120 cm, and 10H-CC. The base of Didymocyrtis tetrathalamus is between Samples 321-U1338A-10H-CC and 11H-3, 110–112 cm. The top of Stichocorys delmontensis is observed between Samples 321-U1338A-11H-3, 110–112 cm, and 11H-5, 73–75 cm. The range of Lychnodictyum anossa is not clearly distinguished in Hole U1338A because of the sparse occurrence of this species.

The evolutionary transition from S. delmontensis to S. peregrina, which defines the boundary between Zones RN9 and RN8, occurs between Samples 321-U1338A-14H-CC and 15H-2, 48–50 cm. The base occurrence of T. vetulum is located between Samples 321-U1338A-15H-2, 48–50 cm, and 15H-4, 97–99 cm. The range of Solenosphaera omnitubus is not clearly distinguished in Hole U1338A because of its rare occurrence.

The top of Diartus hughesi is between Samples 321-U1338A-17H-CC and 18H-2, 68–70 cm. The boundary between Zones RN8 and RN7 is distinguished at this level. The top of Didymocyrtis laticonus also occurs between these two samples. The base of D. penultima and the top of Botryostrobus miralestensis are recorded between Samples 321-U1338A-20H-3, 13–15 cm, and 20H-5, 11–13 cm. The base of S. berminghami is between Samples 321-U1338A-20H-5, 11–13 cm, and 20H-CC.

The top of Zone RN6 is marked by the evolutionary transition of Diartus pettersoni to D. hughesi between Samples 321-U1338A-20H-5, 11–13 cm, and 20H-CC. The top of Stichocorys wolffii was not clearly observed in Site U1338 because of its sparse presence. The base of D. hughesi is recognized between Samples 321-U1338A-20H-CC and 21H-2, 42–44 cm. The base of D. antipenultima is observed between Samples 321-U1338A-22H-5, 53–55 cm, and 22H-CC. The top of Cyrtocapsera japonica is distinguished between Samples 321-U1338A-23H-5, 108–110 cm, and 23H-CC. The top events of both Carpocanopsis cristata and Lithopera neotera are observed between Samples 321-U1338A-24H-2, 42–44 cm, and 24H-5, 133–135 cm. The top occurrences of Cyrtocapsera cornuta, Cyrtocapsera tetrapera, and Lithopera renzae are all located between Samples 321-U1338A-29X-4, 24–26 cm, and 29X-CC.

The base of D. pettersoni occurs between Samples 321-U1338A-31X-CC and 32X-2, 83–85 cm, indicating the top of Zone RN5. The base of L. neotera is located between Samples 321-U1338A-35X-CC and 36X-1, 29–31 cm. The top of Calocycletta robusta is between Samples 321-U1338A-35X-CC and 36X-1, 29–31. Top occurrences of Acrocubus octopyle, Liriospyris parkerae, and Didymocyrtis mammifera are between Samples 321-U1338A-36X-CC and 37X-1, 37–39 cm. Between Samples 321-U1338A-37X-CC and 38X-2, 35–37 cm, top occurrences of Didymocyrtis violina, Calocycletta virginis, and Calocycletta costata are observed. Top events of Carpocanopsis bramlettei and Didymocyrtis tubaria are distinguished between Samples 321-U1338A-39X-2, 75–78 cm, and 39X-5, 87–89 cm.

The Zone RN5/R4 boundary, defined by the evolutionary transition from Dorcadospyris dentata to Dorcadospyris alata, is recognized between Samples 321-U1338A-39X-CC and 40X-1, 103–105 cm. The top occurrence of Liriospyris stauropora is recognized between Samples 321-U1338A-40X-1, 103–105 cm, and 40X-3, 27–29 cm. The base of L. parkerae is distinguished between Samples 321-U1338A-41X-4, 22–24 cm, and 41X-CC. The top of Carpocanopsios cingulata is observed between Samples 321-U1338A-41X-CC and 42X-2, 46–48 cm. The base of Acrocubus octopylus is between Samples 321-U1338A-42X-4, 100–102 cm, and 42X-CC. The base of Zone RN4 is identified by the base of C. costata, recognized between Samples 321-U1338A-43X-2, 63–64 cm, and 43X-CC.

Diatoms

High-resolution biostratigraphy was performed on core catcher and additional samples from Hole U1338A. The diatom stratigraphy at Site U1338 spans the interval from the Fragilariopsis doliolus Zone (late Pleistocene) in Core 321-U1338A-1H (Sample 321-U1338A-1H-CC) to the lowermost part of the Cestodiscus peplum Zone (early Miocene) in the lower part of Core 321-U1338A-42X (Sample 321-U1338A-42X-CC) (Fig. F13; Tables T9, T10). Diatoms are, in general, moderately preserved in the upper 315 m CSF of Site U1338 (Samples 321-U1338A-1H-1, 98–100 cm, through 34X-6, 80 cm; Pleistocene to middle Miocene) and poorly preserved below 316 m CSF (Samples 321-U1338A-34X-CC through 42X-CC; middle Miocene) (Table T10). This decrease in preservation is not clearly associated with any particular lithologic observation, such as the occurrence of turbidites, nor is it associated with reworking of older microfossils. The diverse diatom assemblage is composed of 110 species (Table T10) and, as already described for Site U1337, mainly consists of Neogene species typical of the low-latitude eastern equatorial Pacific Ocean, including Actinocyclus ellipticus, several species of Azpeitia and Coscinodiscus, Nitzschia porteri, Fragilariopsis reinholdii, a few varieties of Thalassionema nitzschioides, and several species of Thalassiosira and Thalassiothrix (Table T10).

The topmost Sample 321-U1338A-1H-1, 98–100 cm, is assigned to the Pleistocene F. doliolus Zone. Samples 321-U1338A-1H-CC through 2H-CC are within the diatom F. reinholdii Zone. The majority of the Pliocene section at Site U1338A is divided into the Rhizosolenia praebergonii and Nitzschia jouseae Zones (Fig. F13). The boundary between the R. praebergonii and F. reinholdii Zones is based on the base occurrence of F. doliolus and the top occurrence of R. praebergonii. The R. praebergonii Zone extends between Samples 321-U1338A-3H-1, 90 cm, and 6H-1, 40 cm. A secondary diatom event in this interval is the top occurrence of N. jouseae in Sample 321-U1338A-4H-CC. The N. jouseae Zone is tentatively defined as the stratigraphic interval from the top Actinocyclus ellipticus f. lanceolata in Sample 321-U1338A-7H-3, 55 cm, and the base occurrence of N. jouseae in Sample 321-U1338A-9H-CC. An additional stratigraphic event observed in this zone is the top occurrence of Fragilariopsis cylindrica (Sample 321-U1338A-9H-6, 110–111 cm).

The expanded Miocene section at Site U1338 is stratigraphically better constrained by diatoms than at Site U1337. The base occurrence of N. jouseae in Sample 321-U1338A-9H-CC defines the top boundary of the Thalassiosira convexa Zone. The lower boundary of the T. convexa Zone is tentatively assigned between Samples 321-U1338A-12H-4, 63–64 cm, and 12H-CC, below the top occurrence of Thalassiosira praeconvexa. The late Miocene Nitzschia miocenica Zone records the base occurrences of T. convexa in Sample 321-U1338A-14H-4, 38–39 cm, T. praeconvexa in Sample 321-U1338A-14H-CC, and Nitzschia miocenica in Sample 321-U1338A-17H-1, 95 cm.

Samples 321-U1338A-17H-4, 55 cm, through 19H-5, 70 cm, are placed in the N. porteri Zone. This zonal placement is based on the top occurrences of Rossiella paleacea in Sample 321-U1338A-17H-5, 55 cm, and Thalassiosira yabei in Sample 321-U1338A-19H-5, 70 cm. A secondary datum event in the N. porteri Zone is the top occurrence of Thalassiosira burckliana (Sample 321-U1338A-18H-3, 65 cm). The N. porteri/T. yabei Zone boundary occurs between Samples 321-U1338A-19H-5, 70 cm, and 19H-6, 70 cm, below the top of T. yabei. The lower boundary of the T. yabei Zone is tentatively assigned to the top occurrence of Actinocyclus moronensis in Samples 321-U1338A-23H-CC and 29H-2, 3–31 cm. The base of A. ellipticus f. lanceolata defines the A. moronensis Zone. The base boundary of the A. moronensis Zone is tentatively allocated between Samples 321-U1338A-25-CC and 26H-2, 25–26 cm. As discussed for Site U1337, the problematic differentiation of the A. moronensis Zone suggests the low biostratigraphic potential of this diatom for the eastern equatorial Pacific Ocean (Baldauf and Iwai, 1995).

The Craspedodiscus coscinodiscus Zone represents the interval between Samples 321-U1338A-26H-2, 25–26 cm, and 30X-CC. Stratigraphic events in this zone are the base occurrences of Hemidiscus cuneiformis (Sample 321-U1338A-26H-CC) and Thalassiosira brunii (Sample 321-U1338A-27X-CC), as well as the top occurrences of Cestodiscus pulchellus (Sample 321-U1338A-29X-CC), and N. porteri (Sample 321-U1338A-30X-CC). The Coscinodiscus gigas var. diorama Zone is placed in Core 321-U1338A-31X.

Samples 321-U1338A-32X-CC through 35X-2, 80 cm, are assigned to the Coscinodiscus lewisianus Zone. This zonal placement is based on the top of C. lewisianus in Sample 321-U1338A-31X-CC and the base of Triceratium cinnamomeum in Sample 321-U1338A-33X-3, 60–61 cm. Samples 321-U1338A-35X-CC through 42X-CC are assigned to the C. peplum Zone. Events occurring in the C. peplum Zone are base occurrences of Thalassiosira tappanae (Sample 321-U1338A-37X-CC) and Actinocyclus ingens (Sample 321-U1338A-40X-1, 122–123 cm). The lower boundary of the C. peplum Zone is tentatively assigned to the top occurrence of Thalassionema fraga in Sample 321-U1338A-42X-CC. Diatom abundance is predominantly low throughout the C. peplum Zone, and valve preservation is moderate to poor. Cores 321-U1338A-43X and 44X are barren.

Planktonic foraminifers

At Site U1338 planktonic foraminifer assemblages are characteristic of tropical-subtropical environments. Assemblages are highly variable and preservation is generally superior to Site U1337. Assemblages are diverse and many of the taxa that were absent in the previously drilled PEAT sites are recorded here.

A planktonic foraminifer biostratigraphy was generated at Site U1338 from Hole U1338A using core catchers and supplemented by additional samples (usually two per core). Six samples were examined from Hole U1338B between Samples 321-U1338B-27H-2, 57–59 cm, and 30H-2, 126–128 cm, to constrain species ranges during an interval of nonrecovery in Hole U1338A. The sedimentary succession at this site ranges from Subzone PT1b (Pleistocene) to Zone M2–M4 (lower Miocene) (Fig. F13; Table T11), which agrees with the calcareous nannofossil, diatom, and radiolarian biostratigraphies except for some minor discrepancies (discussed below). The preservation of planktonic foraminifers varies from poor to very good, though fragmentation and infilling occurs in some samples. Planktonic foraminifer tests generally account for >80% of the total residue in each sample. Taxon-relative abundances and estimates of assemblage preservation are presented in range chart format (Table T12). A number of primary and secondary markers were absent in the Site U1338 samples or had insufficient abundances to provide robust stratigraphic control; these included Globorotalia (Truncorotalia) truncatulinoides, Globorotalia (Hirsutella) cibaoensis, Turborotalita humilis, Globorotalia birnageae, and Catapsydrax dissimilis.

The top of Globigerinoides ruber (pink) is between Samples 321-U1338A-1H-1, 89–91 cm, and 1H-CC, indicating that the top of Hole U1338A is Pleistocene Subzone PT1b and younger than 0.12 Ma. The division between Subzones PT1b and PT1a is indicated by the top of Globorotalia (Truncorotalia) tosaensis between Samples 321-U1338A-2H-2, 70–72 cm, and 2H-5, 115–117 cm. The top of Globigerinoides obliquus is well defined at Site U1338 and is consistently present from Samples 321-U1338A-3H-CC to 20H-2, 56–58 cm, with rare occurrences recorded in Sample 321-U1338A-34X-CC. The top of this species has been calibrated by Chaisson and Pearson (1997) to 1.30 Ma.

The Zone PT1/PL6 boundary is marked by the top of Globigerinoides fistulosus between Samples 321-U1338A-3H-2, 70–72 cm, and 3H-CC. The base of G. fistulosus is between Samples 321-U1338A-5H-2, 117–118 cm, and 5H-CC. The first appearance datum of this species has been correlated with the base of the Mammoth Subchron (Berggren et al., 1995), with a calibrated age of 3.33 Ma for the PEAT timescale. There are a number of secondary markers through Zone PL6. Globigerinoides extremus is present from Samples 321-U1338A-4H-CC to 16H-CC. The top of Globorotalia pseudomiocenica is between Samples 321-U1338A-4H-3, 56–58 cm, and 4H-5, 56–58 cm. As at Site U1337, Globoturborotalita woodi is rare and therefore we were unable to constrain the base of Zone PL6.

Globorotalia (Menardella) multicamerata was present to rare, and we find the top of this taxon between Samples 321-U1338A-4H-5, 56–58 cm, and 4H-CC. Dentoglobigerina altispira and Sphaeroidinellopsis seminulina vary in quantity from present to abundant. The top of both taxa are between Samples 321-U1338A-5H-2, 117–118 cm, and 5H-CC, so (as at Site U1337) we were unable to identify Zone PL4; further samples are required to constrain this zone. We find the top of Globorotalia (Hirsutella) margaritae marking the Zone PL3/PL2 boundary between Samples 321-U1338A-6H-5, 118–120 cm, and 6H-CC. As at Site U1337 this event is hard to determine because the index species is rare.

The change in coiling direction of Pulleniatina spp. from sinistral to dextral occurs between Samples 321-U1338A-6H-5, 118–120 cm, and 6H-CC. The base of Zone PL2, as defined by the top of Globoturborotalita nepenthes, is between Samples 321-U1338A-7H-2, 52–54 cm, and 7H-4, 58–60 cm. Globorotalia (Menardella) exilis is rare, but the base is constrained between Samples 321-U1338A-7H-4, 58–60 cm, and 7H-CC. Globorotalia (Hirsutella) cibaoensis only occurs in two samples from Site U1338; therefore, we are unable to constrain the boundary between Subzones PL1b and PL1a.

Sphaeroidinella dehiscens s.l. occurs from the top of Hole U1338A, with the base of this species between Samples 321-U1338A-8H-2, 43–45 cm, and 8H-5, 106–108 cm. This event is significantly higher than the depth expected by the age-depth plot (Fig. F14). This was also observed at Site U1337 and is perhaps due to the low abundance of the primitive form S. dehiscens f. immatura (Banner and Blow, 1965) at the PEAT sites. Further studies are needed to determine the precise position of the base appearance in this region. Globorotalia tumida is abundant and commonly dominated the samples between the top of Hole U1338A and Sample 321-U1338A-10H-3, 47–49 cm. The base of G. tumida, marking the base of Zone PL1, is between Samples 321-U1338A-10H-3, 47–49 cm, and 10H-6, 108–110 cm. Globoquadrina dehiscens occurs from the base of Hole U1338A to Sample 321-U1338A-10H-CC. This event has been dated at 5.8 Ma (Berggren et al., 1995), and the recalibrated PEAT timescale age is 5.9 Ma. The base of G. margaritae is between Samples 321-U1338A-12H-CC and 13H-2, 105–107 cm; however, as noted above, this species was rare and the age-depth plot (Fig. F14) suggests that this event is depressed by ~10 m. This bioevent has been reported as diachronous from Chron C3An to the base of Chron C3r (Berggren et al., 1995). The age at the PEAT sites will be determined by further studies for documenting its diachroneity. The base of Pulleniatina primalis is between Samples 321-U1338A-11H-5, 65–67 cm, and 11H-CC. This event has been calibrated by Berggren et al. (1995) to 6.4 Ma, with a recalibrated PEAT timescale age of 6.6 Ma. This recalibrated age is generally concordant but somewhat older than that of the age-depth plot of the present site (Fig. F14).

The boundary between Zones PL1 and M14, as marked by the top of Globorotalia lenguaensis, is between Samples 321-U1338A-13H-5, 44–46 cm, and 13H-CC. This species is rare, but the top horizon agrees with the integrated nannofossil, diatom, and radiolarian biostratigraphies at this site. The base of Globorotalia plesiotumida, marking the division between Subzones M13b and M13a, is between Samples 321-U1338A-21H-2, 39–41 cm, and 21H-5, 50–52 cm. This species is distinguished from its ancestral species, Globorotalia merotumida, by having a larger test, a larger increase in chamber-size growth, less convex in edge view on the spiral side, and a final chamber with width narrower than its height. These taxonomic features are clearly observed in specimens at the base horizon. However, the base datum observed at the present site is significantly older than that expected from the age-depth plot (Fig. F14). Further study is required to determine whether these specimens belong to a true evolutional descendant or an ecologic phenotype of the ancestral species. The base of Zone M13, as indicated by the base of Neogloboquadrina acostaensis, is between Samples 321-U1338A-25H-3, 9–11 cm, and 25H-6, 5–7 cm. Paragloborotalia mayeri is abundant and the top of this species is recorded between Samples 321-U1338A-25H-3, 9–11 cm, and 24H-CC. However, we were unable to differentiate Zone M12, as we find the co-occurrence of N. acostaensis and P. mayeri in Sample 321-U1338A-25H-3, 9–11 cm. The absence of Zone M12 (Zone N15 of Blow, 1969) has also been reported in subtropical to temperate successions, namely at North Atlantic ODP Sites 563 and 608 (Berggren et al., 1995), and the Monte Gibliscemi section of Italy (Hilgen et al., 2000). At Site U1337, N. acostaensis is absent below Sample 321-U1337A-18H-3, 141–143 cm, and the top occurrence of P. mayeri could not be defined because of the barren interval from Samples 321-U1337A-18H-CC to 24X-CC. Lourens et al. (2004) proposed the base of regular occurrence of N. acostaensis with an astronomically tuned age of 9.83 Ma. At the present site, it is difficult to recognize the base of regular occurrence because of its sporadic occurrence and low abundance.

Globoturborotalita decoraperta was sporadically present, but the base of this species is between Samples 321-U1338A-27X-CC and 29X-2, 136–138 cm. The top of Globigerinoides subquadratus is also at this horizon. Both the base of G. decoraperta and the top of G. subquadratus are further constrained by analyses in Hole U1338B, where both events are found between Samples 321-U1338B-27H-2, 57–59 cm, and 28H-2, 120–122 cm. G. nepenthes has a sporadic occurrence, and the base of this species (base of Zone M11) is between Samples 321-U1338A-26H-2, 50–52 cm, and 26H-6, 130–132 cm (264.09 m CCSF-A), at a much younger level than expected. However, examination of samples from Hole U1338B indicates that the base of G. nepenthes is between Samples 321-U1338B-27H-2, 57–59 cm, and 28H-2, 120–122 cm (282.42 m CCSF-A), and this occurrence is more consistent with the short duration of Zone M10/N13; therefore, we use this horizon in Figure F13.

Zone M8–M9/N12 is the total range zone of Globorotalia (Fohsella) fohsi. This species is present to abundant, with the top range between Samples 321-U1338A-29X-2, 136–138 cm, and 29X-4, 32–34 cm, and the base between Samples 321-U1338A-35X-2, 9–11 cm, and 35X-CC. The top of the G. fohsi group is supported by analyses from Hole U1338B, with the top between Samples 321-U1338B-28H-2, 120–122 cm, and 28H-6, 107–109 cm. In Sample 321-U1338A-29X-4, 32–34 cm, we find specimens of G. fohsi that are of consistent morphology but diminished in size. This may suggest a dwarfing of G. fohsi in the upper part of its range. Size reduction has also been recognized in other species of planktonic foraminifers prior to their extinction (Wade and Olsson, 2009). Within Zone N12, the base of Gr. (Fohsella) fohsi robusta is between Samples 321-U1338A-34X-2, 78–80 cm, and 34X-4, 91–93 cm, marking the base of Subzone M9b. The top of Globorotalia praescitula is between Samples 321-U1338A-33X-5, 47–49 cm, and 33X-CC. However, the top occurrence is marked by only one specimen and is significantly discontinuous from the lower occurrence. Moreover, the biohorizon is ~20 m higher than that indicated by the age-depth plot of the present site (Fig. F14). There is room for argument whether any bias such as reworking has impacted this datum.

Globorotalia (Fohsella) "praefohsi" is rare. The base of this species, marking the boundary between Zones N10 and N11, is between Samples 321-U1338A-36X-1, 36–38 cm, and 36X-CC. The base of Globorotalia (Fohsella) peripheroacuta, and therefore the base of Zone M7/N10, is between Samples 321-U1338A-36X-CC and 37X-1, 43–45 cm. The top of Globigerinatella insueta is between Samples 321-U1338A-38X-5, 109–111 cm, and 38X-CC. We are unable to constrain the top of Clavatorella bermudezi because of the rarity of this species; however, the base of this taxon can be found between Samples 321-U1338A-37X-CC and 38X-2, 35–37 cm.

The base of Orbulina universa is between Samples 321-U1338A-35X-2, 9–11 cm, and 35X-CC. This species is rare at this site in the lower part of its range, and the base occurrence is much higher than anticipated (Zone M8–M9/N12). We are therefore unable to constrain the base of Zone M6/N9. The division between Subzones M5a and M5b is constrained by the base of Praeorbulina glomerosa between Samples 321-U1338A-40X-CC and 41X-2, 66–68 cm. This interval is also associated with the change in Paragloborotalia from random to exclusively sinistrally coiled, which is between Samples 321-U1338A-40X-1, 115–117 cm, and 41X-2, 66–68 cm (see "Biostratigraphy" in the "Site U1337" chapter for discussion). The Zone M4/M5 boundary is marked by the base of Praeorbulina sicana between Samples 321-U1338A-42X-4, 114–116 cm, and 42X-CC.

The base of G. insueta is between Samples 321-U1338A-43X-CC and 44X-2, 55–57 cm. We are unable to constrain the top of Zone M3 because of the absence of Catapsydrax dissimilis. We find abundant to rare occurrences of Catapsydrax unicavus from the base of the hole to Sample 321-U1338A-29X-4, 32–34 cm (Zone M7/N10). As at Site U1337, we find the top occurrence of Globoquadrina binaiensis (sensu stricto) in an unusually high stratigraphic position. This taxon is present from the base of the hole to Sample 321-U1338A-43X-2, 18–20 cm, above the base of G. insueta. We are therefore unable to differentiate Zones M2 and M3. The extended range of this form requires further investigation. Specimens of Globoquadrina cf. binaiensis (see "Biostratigraphy" in the "Site U1337" chapter for discussion) are also documented from the base of the hole to Sample 321-U1338A-38X-2, 35–37 cm (Subzone M5b).

Microperforate taxa (Globigerinita, Globigerinatella, Mutabella, and Tenuitella) are far more abundant at Site U1338 than at preceding sites. The diversity of this group is particularly high in the interval from Sample 321-U1338A-42X-4, 114–116 cm, to 29X-2, 136–138 cm (Subzone M5a to Zone M10). Stable isotopic evidence suggests that these taxa occupied the mixed layer (Pearson et al., 2001; Majewski, 2003). Our Tenuitella range into Subzone M5b, with a single specimen also recorded from Zone M8–M9/N12, indicating a younger stratigraphic position than previously suggested by Huber et al. (2006) but consistent with Site U1337 and the southern Indian Ocean (ODP Site 744; Majewski, 2003). Between Samples 321-U1338A-42X-4, 114–116 cm, and 29X-2, 136–138 cm, we find variable microperforate globular forms with numerous (often swollen) second generation bullae. Some of these forms may be consistent with Globigerinatella spp. documented from the western tropical Pacific Ocean (ODP Leg 144; Pearson, 1995). Detailed scanning electron microscope analyses are required to determine the full variability in these forms and their relationship to G. insueta.

Within the dentoglobigerinids, we commonly find distinct forms with a flattened umbilical face and incised sutures, referred to here as Dentoglobigerina sp. These forms range from Samples 321-U1338A-41X-CC to 29X-4, 32–34 cm (Subzone M5a to Zone M10/N13) and are abundant in Sample 321-U1338A-39X-2, 72–74 cm. Many of these forms may have previously been considered as Dentoglobigerina larmeui, but further postcruise studies are required to assess the full range of diversity in dentoglobigerinids.

We find distinctive changes in the assemblage composition throughout from the Miocene to Pleistocene (Fig. F16). From the base of Hole U1338A to Sample 321-U1338A-27X-4, 6–8 cm (244.77 CSF; Zone M11) the samples are dominated by species belonging to genera Globoquadrina and Dentoglobigerina. Globoquadrina venezuelana is present from the base of the hole to Sample 321-U1338A-7H-CC and commonly dominates the assemblages. D. larmeui and forms referred to as Dentoglobigerina tripartita are a common component of the assemblage from the base of the hole to Zone M11 and often possess bullae. As at Site U1337, the decrease in globoquadrinids and dentoglobigerinids is associated with an increase in the genus Globigerinoides. Globigerinoides are much more diverse and abundant at this site and species include Globigerinoides conglobatus, Globigerinoides extremus, Globigerinoides fistulosus, Globigerinoides immaturus, Globigerinoides mitra, Globigerinoides obliquus, Globigerinoides quadrilobatus, G. ruber, G. ruber (pink), Globigerinoides sacculifer, Globigerinoides subquadratus, and Globigerinoides trilobus. G. conglobatus is present from Sample 321-U1338A-11H-CC to the top of the hole. Rare to abundant G. immaturus are found between Samples 321-U1338A-33X-CC and 6H-3, 118–120 cm. G. obliquus is present from Sample 321-U1338A-34X-CC but becomes a much more abundant component of the assemblages from Sample 321-U1338A-19H-2, 70–71 cm, to the top of its range in Sample 321-U1338A-3H-CC.

Keeled Globorotalia become a frequent component of the assemblages from Sample 321-U1338A-33X-CC (Zone M8–M9/N12) and become dominant from Sample 321-U1338A-11H-3, 100–102 cm (Zone M14), to the top of the hole (Fig. F16). Globorotalia (Menardella) menardii is abundant from Sample 321-U1338A-22H-2, 51–53 cm, to the top of the hole, with abundant G. tumida from Sample 321-U1338A-10H-3, 47–49 cm, to the top of the hole. Globigerina bulloides is present from the base to the top of the hole. Paragloborotaliids are a distinctive component of the assemblages from the base of the hole until the extinction of P. mayeri (Sample 321-U1338A-25H-3, 9–11 cm). In the upper part of their range (Samples 321-U1338A-32X-4, 67–69 cm, to 25H-6, 5–7 cm), Paragloborotalia siakensis increase their number of chambers in the final whorl from six to seven.

The interval with rare or barren planktonic foraminifers associated with the "carbonate crash" (Lyle et al., 1995; Farrell et al., 1995) ranges from Samples 321-U1338A-22H-5, 41–43 cm, to 24H-CC. This interval is of shorter duration at Site U1338 and restricted to Subzone M13a (Fig. F13), whereas at Site U1337 the interval extends from Zone M11 to Subzone M13a (Fig. F13).

Benthic foraminifers

Benthic foraminifers occur continuously throughout the ~415 m thick Pleistocene to lower Miocene succession recovered at Site U1338, although abundances vary significantly. The early Miocene to Pleistocene assemblages predominantly consist of calcareous taxa with rare agglutinated forms, and the overall composition of assemblages indicates lower bathyal to abyssal paleodepths throughout the Neogene. Prominent variations in the downcore distribution of benthic foraminifers may reflect global climate fluctuations associated with fundamental changes in high-latitude ice volume and Pacific Ocean circulation (Fig. F17). Long-term trends can be correlated across sites drilled during the PEAT expeditions (Sites U1334, U1335, U1337, and U1338), and postcruise studies will provide an opportunity to investigate temporal and spatial variations across the equatorial Pacific Ocean and to unravel links with global climatic and oceanographic events during the Neogene.

Benthic foraminifers were examined in core catcher samples from Hole U1338A, supplemented by samples from Hole U1338A sections (two per core) after cores were split. Mudline samples recovered in Holes U1338A–U1338D were also investigated. Large samples with an average volume of ~50 cm³ were processed from all core catchers to obtain quantitative estimates of benthic foraminifer distribution patterns downcore. Smaller 10 cm³ samples were additionally investigated from core sections to provide realistic estimates of species availability for shore-based geochemical and paleontological studies, although these small samples do not yield statistically significant numbers of specimens. In contrast to Site U1337, the carbonate content of the sediment is generally higher and does not fluctuate so markedly, as prominent intercalations of diatom mats and carbonate-rich layers do not occur throughout the upper Miocene at Site U1338 (see "Geochemistry"). Therefore, sample selection was not biased toward carbonate-rich intervals, and core samples are overall more representative of lithologic variability at Site U1338 than at Site U1337.

To assess assemblage composition and variability downhole, all specimens from the >250 µm fraction were picked from core catcher and core samples and mounted onto slides prior to identification and counting. The distribution of benthic foraminifers was additionally checked in the 150–250 µm fraction to ensure that assemblages in the >250 µm fraction were representative and that small species such as phytodetritus feeders were not overlooked. A total of 88 benthic foraminifer taxa were identified. Census counts from core catcher and core section samples are presented in Table T13. Figure F17 summarizes the downcore distribution of the more common benthic foraminifer taxa in core catcher samples from Hole U1337A. Common taxa include Astrononion echolsi, Cibicidoides mundulus, Cibicidoides grimsdalei, Cibicidoides robertsonianus, Cibicidoides wuellerstorfi, Eggerella bradyi, Fissurina spp., Globocassidulina subglobosa, Gyroidinoides soldanii, Laticarinina pauperata, Oridorsalis umbonatus, Pyrgo murrhina, Pyrgo serrata, Pullenia bulloides, Quinqueloculina spp., and Siphonodosaria abyssorum. Preservation is good overall, but deteriorates slightly in several levels (Cores 321-U1337A-18H through 21H, ~160–190 m CSF, and 26H through 27X, ~239–242 m CSF, and Sections 321-U1337A-35X-2, 37X-1, and 39X-5, ~318.5, ~336.4, ~361.7 m CSF, respectively). Well-preserved fish teeth and ostracodes are intermittently present throughout the succession.

Mudline samples from Holes U1338A–U1338D were gently washed in order to preserve fragile agglutinated specimens with extremely low fossilization potential. All mudline samples examined reveal a moderate degree of dissolution in the planktonic foraminifer assemblage, in particular among globorotalids. The mudline sample from Hole U1338A contains mainly radiolarians with some poorly preserved planktonic foraminifers and rare agglutinated tubular benthic foraminifers. Benthic foraminifers in mudline samples from Holes U1338B–U1338D are generally well preserved, consisting predominantly of agglutinated forms. Rare calcareous taxa include Astrononion echolsi, Cibicidoides pachyderma, C. wuellerstorfi, Fissurina formosa, Fursenkoina bradyi, G. subglobosa, G. soldanii, L. pauperata, Melonis pompilioides, Melonis sphaeroides, Pyrgo spp., and Quinqueloculina spp. Agglutinated taxa are mainly tubular forms with organic cement, such as the branching species Rhizammina algaeformis, which typically agglutinates a variety of planktonic foraminifer tests. Also present are the coarsely agglutinated Reophax dentaliniformis, Lagenammina diffugiformis, and Psammosphaera sp. and finely agglutinated Buzasina glaeata, Hyperammina elongate, and the calcareous agglutinated species E. bradyi. The sample from Hole U1338B contains exceptionally well preserved large specimens of Astrorhizza arenaria not found in other mudline samples. The sample from Hole U1338D contains two well-preserved tests of the planktonic species Beella digitata, which rarely fossilizes. Highly fragile agglutinated assemblages typical of lower bathyal to abyssal depths occur in all mudline samples but are best preserved in Holes U1338B–U1338D, suggesting sediment recovery close to the sediment/water interface.

The Pleistocene to Pliocene assemblage within the biosiliceous-rich interval in the upper part of Hole U1338A (Cores 321-U1338A-1H through 8H; 2.68–69.65 m CSF) contains relatively low numbers of benthic foraminifers, representing only >1% of the total planktonic and benthic foraminifer assemblages. Variations in the abundance of organic flux–sensitive taxa such as G. subglobosa, O. umbonatus, P. murrhina, and P. serrata suggest that changes in equatorial Pacific Ocean surface productivity occurred throughout the Pleistocene to Pliocene, possibly associated with glacial–interglacial climate fluctuations. Sample 321-U1338A-6H-CC (50.70 m CSF), within an interval of lower CaCO₃ values (see "Geochemistry"), is characterized by high numbers of infaunal taxa (25%) including Siphodonosaria abyssorum and Siphodonosaria spp., which suggests that episodic decreases in bottom water ventilation occurred during the Pliocene. However, the low resolution of our shipboard data set prevents detection of variability that may occur on orbital to suborbital timescales.

Benthic foraminifer abundance increases and the benthic/planktonic ratio oscillates markedly during the latest Miocene (Cores 321-U1338A-10H through 13H; 97.05–126.80 m CSF). Sample 321-U1138A-14H-CC (126.80 m CSF), within a diatom-rich interval, is effectively barren of foraminifers. A prominent increase in miliolids is noted in Sample 321-U1338A-11H-CC (97.05 m CSF), where this group represents 37% of the total assemblage. The late Miocene assemblage in Cores 321-U1338A-14H through 21H (126.80–192.95 m CSF) generally shows high abundance and diversity, including high numbers of epifaunal or near-surface dwellers including A. echolsi, C. mundulus, C. wuellerstorfi, G. soldanii, G. subglobosa, L. pauperata, O. umbonatus, P. murrhina, P. serrata, and Quinqueloculina spp. The lowest occurrence of Osangularia plummerae in Hole U1338A occurs within this interval (Sample 321-U1338A-20X-5, 56–58 cm; 180.26 m CSF). Fluctuations in assemblage composition and benthic/planktonic ratio are evident, hinting at major changes in upper and deep ocean structure, which cannot be captured by our low-resolution data set. This interval coincides with improved carbonate preservation at the seafloor following an extended period of low carbonate deposition, widely recorded at ~9–11 Ma throughout the eastern Pacific Ocean and often referred to as the carbonate crash (Lyle et al., 1995; Farrell et al., 1995). This transition to higher carbonate deposition marks a fundamental change in Pacific Ocean circulation and productivity regime, possibly associated with changing rates of deep water production and/or Northern Hemisphere ice development (Zachos et al., 2008).

Core catcher Samples 321-U1338A-22X-CC through 27X-CC (202.58–247.60 m CSF) are either barren or contain low numbers of planktonic foraminifers and are generally characterized by an impoverished benthic foraminifer assemblage. Sample 321-U1338A-23X-CC (212.16 m CSF), within a diatom-rich section, is effectively barren of benthic foraminifers. This interval corresponds to the early late Miocene carbonate crash and was also recorded in Site U1337. Interestingly, this prominent dissolution event appears less intense at Site U1338 than at Site U1337, as shown by higher numbers of planktonic and benthic foraminifers and higher carbonate content throughout (see "Geochemistry") at Site U1338. This may be due to the fact that the site was located on younger oceanic crust and was closer to the Equator during this time window.

The middle to early Miocene assemblage in Cores 321-U1338A-29X through 41X (268.33–382.72 m CSF) is overall diverse and abundant, particularly within Cores 34X and 42X (316.10–391.07 m CSF). However, some fluctuations in abundance and composition are evident, which may reflect cyclic changes in surface productivity and circulation but cannot be ascertained because of the low resolution of our samples. The assemblage is characterized by epifaunal species such as C. mundulus, C. wuellerstorfi, and L. pauperata and by mobile, shallow infaunal dwellers living close to the sediment/water interface, such as Cibicidoides bradyi, C. grimsdalei, C. robertionianus, G. soldanii, G. subglobosa, O. umbonatus, and Pyrgo spp. The benthic/planktonic ratio remains relatively low (~1%–20%) but rises intermittently within this interval, suggesting transient variations in carbonate preservation linked to changes in the CCD. Enhanced carbonate preservation at the seafloor during this interval is supported by overall high CaCO₃ values (see "Geochemistry"). This interval appears to coincide with the prolonged period of increased CaCO₃ deposition, widely identified at ~13–16 Ma in the eastern equatorial Pacific Ocean (Lyle, 2003). Interestingly, most of this interval corresponds to the episode of major global cooling at ~13–15 Ma, when ice sheets expanded in Antarctica toward the end of the "Monterey carbon isotope excursion" (Lewis et al., 2007). A substantial improvement in deepwater ventilation and deepening of the calcium carbonate compensation depth was also recorded at ODP Site 1237 in the southeastern subtropical Pacific Ocean following middle Miocene ice growth (Holbourn et al., 2005; 2007).

The abundance and diversity of the early Miocene assemblage in Cores 321-U1338A-42X through 44X (391.07–408.34 m CSF) are generally lower, but marked variations in assemblage composition occur, indicating probable changes in water mass characteristics which are beyond the resolution of our sample set. For instance, Sample 321-U1338A-43X-2, 18–20 cm (394.88 CSF), is characterized by high numbers of infaunal taxa (>40%), including S. abyssorum and Siphonodosaria spp., and exhibits a high benthic/planktonic ratio (30%), which suggests somewhat reduced bottom water ventilation. In contrast, some of the other lower Miocene samples contain higher numbers of epifaunal taxa and have much lower benthic/planktonic ratios (1%–7%). This interval, where benthic foraminifer distribution remains highly variable, corresponds to a period of global warmth during the early part of the Monterey carbon isotope excursion, often referred to as the mid-Miocene climatic optimum. Sample 321-U1338A-44X-CC (408.34 CSF), which is close to the basaltic basement, contains heavily recrystallized benthic foraminifers.

Top of page   |   Previous   |   Next