With two-thirds of the cores taken by APC as well as good sediment recovery via rotary coring, Leg 206 has added considerably to the sediment recovery for this region. The high core recovery (89%) provides an excellent opportunity to study environmental, oceanographic, and biotic changes from the middle Miocene to the Holocene in the equatorial Pacific. Sedimentation rates especially were generally high in those critical time intervals, which is ideal for high-resolution biostratigraphic and paleoceanographic studies.
Linear sedimentation rates, uncorrected for compaction, were calculated based on 28 nannofossil age estimates in Table T2 for the sedimentary sequence of Hole 1256B and plotted with best-fit lines. Sedimentation rates vary from 6.1 to 39.1 m/m.y. (Fig. F2A). Calculated sedimentation rates are high in the middle Miocene (39.1 m/m.y.), decrease drastically in the lower upper Miocene (8.4 m/m.y.), recover somewhat in the uppermost Miocene (13.2 m/m.y.), and reach the lowest point in the Pliocene (6.1 m/m.y.), but they increase thereafter (11.8 m/m.y.).
In order to illustrate changes in depositional environments and processes through time at Site 1256, mass accumulation rates (MARs) have been calculated for bulk sediments and the main sedimentary components such as calcium carbonate, terrigenous material, and biogenic silica (Fig. F3).
Calcium carbonate MARs show an abrupt transition at ~11 Ma, which coincides with a "diatom mat" (i.e., a series of diatom-rich laminae that occur at this horizon throughout the region; see Kemp and Baldauf, 1993). Carbonate MARs are high except for a major drop at ~12.9 Ma during the middle Miocene, remain extremely low from ~10.2 to 9.1 Ma, recover somewhat from 9 to ~5 Ma, and stay particularly low thereafter.
This pattern of variations in sedimentation was also observed at Sites 844–853 drilled during ODP Leg 138 (Fig. F1). Although there is a difference in proximity to continents and in water depths among sites located in Guatemala Basin (Sites 844, 845, and 1256), sedimentation rates positively covary (Fig. F2B), and sedimentary thickness during a specific time interval shows little variation (Fig. F4). This suggests that these sites have experienced a virtually same sedimentation history and terrestrial influences have remained a minor factor from the middle Miocene through the Quaternary. This, together with the fact that positions of the epoch boundaries are closely related to water depths (Fig. F4), reflects virtually no intraplate variation in geological and tectonic settings on Cocos plate.
The nature of deposition in EEP has been influenced mainly by surface water productivity. Sites on the Cocos and Nazca plates are separated from the west coast of Central America by the Middle America Trench, which traps most terrestrial sediments shed from the continent, and sites on the Pacific plate are far removed from continents. The sedimentary sources, therefore, are mainly biogenic calcite and silica, which determine the sedimentation rates (e.g., at Site 1256 [Fig. F3]). This is further confirmed by the close correlation between CaCO3 MARs and Ba/Ti ratios, chemical proxies for biogenic productivity (Shipboard Scientific Party, 2003).
The distribution pattern of surface water productivity today helps explain the spatial and temporal variations in EEP. The V-shape profile in the longitudinal transect along Sites 848–853 clearly reflects diminishing productivity away from the Equator (Fig. F4), as seen today (Paytan et al., 1996). Sites with substantially higher sedimentation rates are located under highly productive waters (e.g., Sites 849–851 under the equatorial divergence and Sites 846–847 close to the Galápagos Islands under strong influence of the Peru upwelling system). These sites have been remaining under this situation since middle Miocene because of a moving path nearly parallel to the Equator (Pisias et al., 1995). The high biological productivity fuels the high sedimentation rates. Conversely, the diminishing productivity has resulted in a decrease in sedimentation rates as sites on the Cocos plate (Sites 844, 845, and 1256) move away from the Equator.
The high sedimentation rates in the middle Miocene can be attributed to high productivity and good preservation while these sites were near the paleoequator on young, shallow seafloor. This is indicated by a nannofossil assemblage showing little dissolution at Site 1256. The lowest sedimentation rates (Pliocene), however, likely have resulted from intense dissolution superimposed on reduced paleoproductivity as indicated by the poor to moderate preservation of nannofossils and nannofossil-barren intervals at Site 1256. This probably resulted from a combination of seafloor subsidence and a shoaling of the calcite compensation depth (CCD) associated with a change in chemistry of deep waters and/or surface carbonate production.
The major middle–late Miocene carbonate shift recorded at Site 1256 is referred to as the carbonate crash by Lyle et al. (1995) and Farrell et al. (1995). This event has been widely recorded in the Pacific and Atlantic Oceans and Caribbean Sea (Berger, 1970; King et al., 1997; Roth et al., 2000) and provides a seismic reflector for long-range correlation (Mayer et al., 1986; Bloomer et al., 1995). Lyle et al. (1995) and Farrell et al. (1995) attributed the carbonate crash to enhanced dissolution associated with changes in bottom water characteristics, rather than with changes in productivity. Emerson and Bender (1981) and Archer (1991a, 1991b), however, have demonstrated that under proper conditions, increases in productivity can lead to enhanced dissolution through acids produced by the degradation of organic matter. Lyle et al. (1995) cited a decrease rather than increase of organic carbon to support their argument, which seems plausible because such a decrease in organic carbon can also be caused by enhanced degradation resulting in higher acid production. In such a case, the organic carbon cannot be used as a proxy of productivity.
Several mechanisms have been proposed for the carbonate crash. Even though there is an apparent correlation of the carbonate crash and a eustatic sea level drop (Haq et al., 1987), Peterson et al. (1992) showed that globally, the middle–late Miocene sea level drop is associated with deepening of the CCD and enhanced carbonate preservation. This is supported by Berger's (1970) basin-shelf fractionation model. Other causative mechanisms proposed, such as the constriction of the Panama Gateway (Lyle et al., 1995) and the initiation of North Atlantic Deep Water (NADW) (Woodruff and Savin, 1989), account for the crash by enhanced corrosiveness of bottom water imported into this region.
The arguments presented above clearly show that the middle–late Miocene carbonate crash is still not resolved in terms of the interplay of dissolution and production. Dissolution seems to play a certain role associated with large variations in the CCD, with nearly complete carbonate dissolution during brief intervals when the CCD shoaled by as much as 1400 m (van Andel et al., 1975). However, the cause of enhanced dissolution is complex but probably relates to (1) tectonic movement associated with the Panama Gateway and a slowdown of seafloor spreading, as indicated by the magnetic anomaly pattern in Figure F1; and (2) phytoplankton productivity associated with ocean circulation patterns.
Evidence from calcareous nannofossils and diatoms tends to attribute it to surface water process. Prior to the carbonate crash, all major reductions in carbonate resulted from deposition of laminated diatom oozes (Kemp, 1995), which is believed to be controlled by surface process, instead of dissolution at depth (Kemp et al., 1995). At Site 1256, the calcareous nannofossil assemblages exhibit their best preservation just prior to the carbonate crash nadir. The same is true at Site 846 (Raffi and Flores, 1995). This proposal is consistent with the study in terms of stable isotopes (Jiang et al., this volume).
Calcareous nannofossils are susceptible to reworking because of their tiny size; hence, reworked fossils often pose problems in biostratigraphy. Some of them with a short range but that are clearly out of place (e.g., D. hamatus), together with lithologic information, however, provide an opportunity to study episodes of redeposition. For example, downhole investigation of nannofossils found D. quinqueramus and D. hamatus together in Samples 206-1256B-9H-4, 40–42 cm (77.50 mbsf), and 9H-4, 115–117 cm (78.25 mbsf), which caused confusion in interpreting the biostratigraphy. Upon close examination of more samples, the isolated occurrences of D. hamatus in these two samples seem more likely to be reworked, and hence its actual LO should be much lower, which is supported by the sudden sedimentary color change. Based on the criteria of reworked fossils and abrupt lithologic change, three episodes of redeposition were recognized, and the timing for these were obtained from the age-depth plot assuming constant sedimentation rates for the specific interval:
These redepositional events provide cogent evidence for downslope processes active on the EPR throughout Neogene. It is worth noting that Episode 2 has been recorded in Knüttel's 1986 study on the EPR as well as in many ocean basins (Ciesielski and Wise, 1977; Adams et al., 1977). Knüttel (1986) suggested that a Messinian sea level drop triggered intensified bottom current activity, whereas Rea and Janecek (1986) alternatively interpreted this as tectonic instability associated with possible increased seismic activity. By either mechanism, however, turbidity currents have produced a rapid accumulation of sediments as indicated by the sharp sedimentary contacts and sharply reduced bioturbation.