A synthesis of the occurrence of calcareous nannofossils recovered from the cored sequences at Sites 1039, 1040, 1041, 1042, and 1043 is presented. For each site, a composite section combining all of the holes cored has been constructed and the geologic framework for each site is characterized. Because the primary purpose of the nannofossil analyses at all the Leg 170 sites is to establish a geologic age-depth model for the sediments cored at each site, most of the nannofossil analyses centers on locating the top and bottom of the ranges of index nannofossil species in the cores and assigning nannofossil zones to the cored sequences. The zonation of Martini (1971) provides the nannofossil zonation framework for this study. Martini's zones are defined in Table T8 and are determined by the range of the first and/or last occurrences of the listed index species. Also defined in Table T8 is the widely used nannofossil zonation scheme of Bukry (1973b, 1975) and Okada and Bukry (1980), an alternative to Martini's zonation scheme that can easily be adapted to the Leg 170 nannofossil data.
Three holes were cored at Site 1039 (located along seismic line CR-20) in a water depth of ~4352 m through the trench floor on the Cocos plate (Figs. F2, F4). Hole 1039A penetrated 28 meters below seafloor (mbsf) and recovered three cores. Hole 1039B penetrated 384.3 mbsf and recovered 42 cores. Hole 1039C was drilled to 363.1 mbsf and then cored to a depth of 448.7 mbsf; 11 cores were recovered. A composite diagram of total core recovery, depth cored, lithology, and age is provided in Figure F7. Core recovery from Holes 1039A and 1039B is near 100%, enhancing biostratigraphic resolution; core recovery from Hole 1039C is 44% primarily due to difficulties in drilling through 85 m of soft ooze, well-indurated breccia, and gabbro.
The upper 84 m (lithologic Unit U1) of sediment at Site 1039 is primarily a carbonate-poor diatom ooze (Fig. F7). In this section, calcareous nannofossils are present in only about one-third of the intervals sampled. The top 5.5 m of this section is comprised of turbidite.
A sharp decrease in both calcareous nannofossil and diatom abundances is observed between 84 and 133 mbsf (lithologic Unit U2). These sediments are silty clays near the top, but they become increasingly more calcareous downsection (133-152 mbsf) (Fig. F7). Pore-water geochemistry between 95 and 130 mbsf indicates low chloride concentrations (545 nM) corresponding to a 2% seawater dilution (Kimura, Silver, Blum, et al., 1997). According to Kimura, Silver, Blum, et al. (1997), this interval probably indicates a past low-chloride fluid conduit. Based on the diffusivity of chloride into the surrounding fluids, they suggest that this conduit existed prior to 180 ka. It is further suggested that the fluid flowing in this conduit prior to 180 ka affected the preservation of calcareous nannofossils and fossil diatoms. Nannofossil-size micrite is very abundant in the nannofossil-barren intervals between 95 and 130 mbsf and may represent diagenetically altered nannofossils. Subunit U2A (84.43-152.49 mbsf) contains <5% carbonate, whereas Subunit U2B has a fluctuating carbonate content of 5% to >20%. Subunit U2B represents a transitional zone from the hemipelagic sediments above to siliceous nannofossil ooze at 152 mbsf. Subunit U3A (152-180.38 mbsf) sediments are predominately siliceous nannofossil ooze with horizons of nannofossil-rich clay. Micrite is very common throughout these units, and carbonate content averages between 20% and 40%.
Subunit U3B (180.38-279.93 mbsf) consists of nannofossil ooze with >50% carbonate content. The change between Subunits U3A and U3B appears to be transitional until the abrupt disappearance of the clay layers at 180.38 mbsf. Nannofossils, diatoms, sponge spicules, foraminifers, radiolarians, and silicoflagellates (in order of decreasing abundance) make up the bulk of the sediment grains of Subunit U3B (Kimura, Silver, Blum, et al., 1997). Micrite is ubiquitous, and terrigenous grains of quartz and feldspar are absent or sporadically present in only trace amounts.
The siliceous component increases in proportion to the carbonate component beginning at 280 mbsf with the introduction of thin layers of diatom oozes consisting of mats of a single species of diatom Thalassiothrix longissima. Similar mats were observed during Leg 138 directly west of Site 139 in the eastern equatorial Pacific (Kimura, Silver, Blum, et al., 1997). Another characteristic of this lithology is the presence of coarse-grained matrix-supported breccias with clasts of carbonate and siliceous ooze. Analysis of the nannofossil assemblage of one of these clasts points to a late Miocene age for the assemblage. Subunit U3C extends to 378 mbsf.
Gabbro (basement?) was first encountered at 379 mbsf in Hole 1039B and at 422 mbsf in Hole 1039C (lithologic Unit U4). Mineralogically, samples from both sites seem to indicate the same intrusion (Kimura, Silver, Blum, et al., 1997). If the gabbro represents true basement rock, then the constrained age of the basement at this site is some 8 to 9 m.y. younger than previously thought. This will be discussed in detail in the next section.
The distribution, abundance, and preservation of calcareous nannofossils recovered from Holes 1039A, 1039B, and 1039C correlated to the nannofossil zonation scheme of Martini (1971) are presented in Table T1, in PDF format. An age-depth diagram plotting the age of all observed nannofossil datums at Site 1039 with the depth of observation is presented in Figure F8. The shipboard-derived paleomagnetic dates plotted in Figure F8 strongly match the age-depth relationships of the calcareous nannofossil datums in the Pleistocene and Pliocene intervals. However, within the lower Miocene, there is clearly an abnormality in the paleomagnetic data, which indicates sediment accumulation rates much higher than normally expected for pelagic sediments.
In the top 84 m of the Site 1039 composite section, calcareous nannofossil preservation is poor to moderate and species occur very sporadically with depth (Table T1, in PDF format). Nannofossil Zones NN21 and NN20 are observed. The bottom of Emiliania huxleyi occurs at a depth of 31.03 mbsf and represents an age of 260 ka (Berggren et al., 1995b). The top of Pseudoemiliania lacunosa occurs at a depth of 48.75 mbsf and represents and age of 460 ka. The error associated with this depth, regarding the actual top of P. lacunosa at this site, is -3 m because of barren intervals immediately upsection.
Nannofossils are essentially barren in samples taken between 84 and 120 mbsf (Table T1, in PDF format). As discussed earlier, the possibility of a fluid conduit between 95 and 130 mbsf before 180 ka may have contributed to the poor preservation of calcareous nannofossils in that interval. Within this section, however, are three locations that do contain nannofossils in sufficient quantity for biostratigraphy. One sample at 113 mbsf contains abundant nannofossils, albeit low in diversity, associated with micrite and with some evidence of etching. Another at 105 mbsf contains few poorly preserved species and associated micrite, and a third sample, at 97.37 mbsf, has common species abundances with moderate preservation and micrite. These samples have provided sufficient amounts of nannofossils to delineate two tops, Helicosphaera sellii at 97.37 mbsf and Discoaster brouwerii at 105.5 mbsf. The LOD of H. sellii is 1.47 Ma; the LOD of D. brouwerii is 1.95 Ma. The LOD of D. brouwerii to the LOD of P. lacunosa defines nannofossil Zone NN19.
The top of Discoaster pentaradiatus at 120.37 mbsf, corresponding to an LOD of 2.55 Ma, is found at the top of a 7-m sequence of abundant moderately to well-preserved nannofossils but still within the section represented by generally poor preservation or barren intervals due to the possible low-chloride pore-water conduit. The locations of the tops of H. sellii, D. brouwerii, and D. pentaradiatus in the cores (see Table T1, in PDF format) seem to indicate that their actual tops possibly may have been several meters upsection had the overlying intervals not been barren. However, when plotted against the paleomagnetic record for Site 1039, the top of each of these species integrates nicely with the paleomagnetic record (Fig. F8). Thus, the tops of these three species appear to represent highly reliable datums. The top of D. pentaradiatus to the top of D. brouwerii defines nannofossil Zone NN18. The tops of Reticulofenestra pseudoumbilica and Sphenolithus species are found at 134.21 mbsf. Both have a LOD of 3.75 Ma. Their tops in this core are considered to be reliable also.
Missing at Site 1039, between the tops of D. pentaradiatus and R. pseudoumbilica, are the tops of D. surculus (LOD = 2.55-2.59 Ma) and D. tamalis (2.73-2.78 Ma) (Table T1, in PDF format). Because these datums are not observed at Site 1039, the calcareous nannofossil zones defined by these datums (Zones NN17 and NN16 of Martini [1971] and Subzones CN12a, CN12b, and CN12c [Bukry, 1973a, 1973b]) cannot be resolved. However, the age-depth diagram for Site 1039 (Fig. F8) does not seem to indicate any breaks in the sediment age-depth record. Even more interesting, these two datums are not observed at any of the other sites cored during Leg 170 (see "Site 1040" and "Site 1043"). Evidence for a hiatus at these intervals is not supported by the Site 1039 age-depth curve (Fig. F8). It is possible then that both D. surculus and D. tamalis simply were not present in the area before and up to their last appearance datums of 2.55 and 2.78 Ma, respectively. Similarly, at Deep Sea Drilling Project (DSDP) Site 155 in the Panama Basin, Bukry (1973a) was unable to find either D. surculus or D. tamalis. He did, however, resolve the time interval represented by the D. tamalis subzone (CN12a) utilizing the occurrence of Thoracosphaera saxea and the absence of Reticulfenestra pseudoumbilica.
Calcareous nannofossil diversity and abundance increase markedly and preservation quality improves from 140 mbsf to the base of the hole at 422 mbsf (Table T1, in PDF format). This corresponds to the nannofossil ooze interval discussed previously in this section. Thirteen datums are observed in this interval (Table T1, in PDF format). All thirteen datums are considered very reliable. They all occur in well-preserved, abundant intervals with high species diversity. Each index fossil is abundant when it is encountered, and each follows the other in a very predictable sequence.
Five expected datums were not observed between 140 and 422 mbsf; thus, the zones that are defined by these datums cannot be delineated. The bottom and top of Amaurolithus primus, corresponding to 7.2 and 4.8 Ma, respectively, were not observed. Amaurolithus primus was observed in one sample only at ~1 specimen per 100 fields of view (rare). Amaurolithus amplificus (LOD = 5.9 Ma) was observed in only one sample, and its abundance is recorded as rare. Likewise, the zones defined by these datums have been combined (Table T1, in PDF format) because the individual zones in this interval (NN16, NN15, NN14, and NN13) cannot be resolved. At DSDP Site 155, Bukry (1973a) also faced similar problems in resolving these zones because of the absence or scarcity of the marker species.
Catinaster calyculus and Catinaster coalitus, with corresponding FODs of 10.7 and 11.3 Ma, respectively, were not observed at Site 1039 or at any other Leg 170 site (the FOD of C. coalitus defines the base of CN6 or NN8; thus, Zones NN7 and NN8 could not be differentiated). Another species, Calcidiscus macintyrei, with an LOD of 1.59 Ma and an FOD of 15.6 Ma was also not observed. For this study, one of the characteristics adopted for calcidiscids is from E. DeKaenel (pers. comm., 1998), who has shown that there is a bimodal distribution of the Calcidiscus species and that to be considered C. macintyrei the specimen must be at least 11 µm in diameter. Only a few Calcidiscus species in the Leg 170 cores had a greater diameter; none was a top or bottom that represented a first or last occurrence.
The base of the cored sequence at Site 1039 consists of gabbro intruded into a calcareous diatomaceous ooze and breccia. Gabbro is present with the ooze in the basal core of Hole 1039B and in Cores 170-1039C-2R through 7R (Fig. F7). Reliable age-diagnostic calcareous nannofossils (Sphenolithus heteromorphus and Helicosphaera ampliaperta) are present in all samples taken from those cores (Table T1, in PDF format). They are reliable because they occur in a thick interval of consistently very abundant nannofossils (offering high stratigraphic resolution and no barren intervals) that exhibit good preservation. If the gabbro is indeed basement rock, then the age must be between 15.6 and 18.2 Ma. Hey (1977) and Lonsdale and Klitgord (1978) have suggested that the age of the Cocos plate entering the subduction zone is between 25 and 27 Ma. However, Wilson (1996) has implied a younger age for the subducting lithosphere (Kimura, Silver, Blum, et al., 1997). Sediments examined during Leg 138, ~400 miles west of Site 1039, suggested ages of 15-17 Ma for the basement at Site 844. If that is the case, then at average East Pacific Rise spreading rates, the age of basement at Site 1039 could be between 21 and 25 Ma.
Site 1040 is located along seismic line CR-20 on the landward side of the Middle America Trench on the Caribbean plate at a water depth of ~4178 m, which is about 3.13 km from Site 1039 and ~1.6 m upgradient from the toe of the slope (Figs. F1, F2, F4). Three holes were cored through a deformed hemipelagic sedimentary wedge and the décollement and into the same three hemipelagic and pelagic sediment units and one oceanic basement unit cored at reference Site 1039 (Fig. F7) on the Cocos plate that has underthrust the Caribbean plate at Site 1040. A total composite penetration of 665 mbsf was achieved.
A composite diagram of total core recovery, lithology, depth cored, and age is illustrated in Figure F7. Hole 1040A penetrated 9.5 m of sediment recovered in one core. Hole 1040B penetrated to 190.2 mbsf, recovering 22 cores at a recovery rate of 67.3%. Hole 1040C was drilled to 159.3 mbsf before coring began. Fifty-three cores were taken from Hole 1040C, penetrating 505.7 m below 159.3 mbsf. Total core recovery at Hole 1040C was 74.6%.
Four sedimentary units and one igneous lithologic unit were recovered at Site 1040 (Fig. F7). Those units that make up the sedimentary wedge extend from 0 to 371 mbsf and are designated with a "P" for prism. From 371 to 653 mbsf, the same lithologic units cored at Site 1039 are present and are designated "U" for underthrust.
Lithologic Unit P1 extends from the seafloor to a depth of 371 mbsf. This unit comprises all of the prism section overlying the underthrust sections. Carbonate sediments and nannofossils are rare in Unit P1, as are diatoms, foraminifers, and radiolarians. Unit P1 consists primarily of massive silty clay to claystones interbedded with a few thin sandy layers. Slump features and breccia within the upper part of Unit P1 are suggestive of mass downslope transport processes. The lack of clear grading or bedding, the presence of slump folds, and the few bedded sands (turbidites) in the prism sediments indicate that most, if not all, of the prism has been mixed and reworked. This, of course, increases the difficulty in dating the prism sediments with fossils. The décollement is recognized at Site 1040 as a shear zone between Cores 170-1040C-19R through 23R. The bottom of the décollement is sharply defined at a depth of 371 mbsf. The top of the décollement is uncertain due to poor core recovery.
Beneath the décollement, lithology changes sharply from the massive silty clays and claystones characteristic of the prism to the clayey diatomite of lithologic Unit U1 first delineated at Site 1039. Figure F7 compares the underthrust units at Site 1039 (before subduction) to the same units at Site 1040 (after subduction). The lithology of the underthrust lithologic units are discussed in detail in "Site 1039" and will not be repeated here. One of the few features found in the Site 1040 underthrust section not seen in the Site 1039 cores is the presence of small-scale faults and intervals of significant dip, indicating small-scale folding. The major difference in the lithologic sequence between the two sites is in the thickness of the corresponding lithologic units. Lithologic Units U1 and U2 at Site 1040 are ~67% of the thickness of the corresponding units at Site 1039. Unit U3 at Site 1040 is ~80% of the thickness of corresponding Unit U3 at Site 1039.
Interestingly, the base of the sedimentary section U3C, located just above the gabbro intrusion, exhibits an increase in deformation. Shipboard structural geologists believe that this deformation is caused by near-ridge tectonics "very early in the sediment history" (Kimura, Silver, Blum, et al., 1997).
The distribution, abundance, and preservation of calcareous nannofossils recovered from Holes 1040B and 1040C correlated to the nannofossil zonation scheme of Martini (1971) are presented in Tables T2, in PDF format; and T3, in PDF format. An age-depth diagram plotting the age of all observed nannofossil datums at Site 1040 against the depth of observation is presented in Figure F9.
Relatively thin intervals of mixed Pleistocene, Pliocene, and middle and late Miocene calcareous nannofossil assemblages are characteristic of the 371-m-thick silty clay wedge section at Site 1040. Approximately 72% of the intervals analyzed in the upper 303 m of Unit P1 are barren. The longest nannofossil-barren interval, from 180.7 to 215.14 mbsf, coincides remarkably well with a zone of low chloride content in the pore waters. This zone, centered at 200 mbsf, contains pore waters that are up to 29% more diluted (chloride = 400 mM) than normal seawater (Kimura, Silver, Blum, et al., 1997), indicating that a conduit of low-chloride fluid advected from deeper in the wedge. It is likely that this low-chloride pore-water conduit has contributed to the absence of nannofossils and the paucity of diatoms characteristic of this horizon. In fact, comparison of the Holes 1040B and 1040C nannofossil distributions (Tables T2, in PDF format; T3, in PDF format) with the concentration of chloride with depth profiles in figure 24 (pg. 131) of Kimura, Silver, Blum, et al. (1997) indicates that most, if not all, of the nannofossil-barren zones at this site are associated with low-chloride concentration pore waters.
Within the nannofossil-bearing intervals, Pliocene through middle Miocene nannofossil reworking is ubiquitous (Table T2, in PDF format). Because index nannofossils observed from the wedge are rare or absent and because of the intense degree of reworking, all wedge age assignments must be considered conditional.
Careful study of the relatively few nannofossil-bearing intervals in Unit P1 indicates that from 0 to 14.25 mbsf (Section 170-1040B-3X-CC), the sediments are late Pleistocene in age based on a nannofossil assemblage characteristic of calcareous nannofossil Zone NN21 (Table T2, in PDF format). Zone NN21 is defined as extending from the FOD of E. huxleyi (260 ka to present). Thus, between 0 and 14.25 mbsf, the sediments are tentatively considered to be younger than 260 ka. From 14.25 to 303.91 mbsf (Section 170-1040C-15R-CC), the nannofossil assemblages cannot be zoned, although the assemblages are also characteristic of Pleistocene assemblages (Zones NN20 or NN19).
D. brouweri is first seen at 305.68 mbsf in Sample 170-1040C-16R-2, 48-50 cm, working downsection (Table T3, in PDF format). The LOD of D. brouweri defines the top of nannofossil Zone NN18 and also the top of the Pliocene. Although its abundance is rare in this sample and to a depth of 317.58 mbsf, this interval is tentatively assigned to late Pliocene nannofossil Zone NN18. Within the deformed sedimentary wedge (0-330 mbsf), numerous normal and reversed polarity intervals have been determined. Lack of good biostratigraphic control and the high probability of thrust faults within the deformed sedimentary wedge have so far prevented a unique assignment of the reversal stratigraphy to any specific portion of the section.
The décollement at Site 1040 lies between Samples 170-1040C-19R-2, 37-39 cm, (334 mbsf) and 22R-CC (371 mbsf) (Table T3, in PDF format). Within this 37-m interval, only 23% of the total range contains nannofossils. The nannofossil assemblages in the décollement are very poorly preserved and rare to very few in number. Based solely on the nannofossil assemblage, this interval cannot be dated.
Directly underlying the décollement at Site 1040 is the underthrust extension of lithologic Unit U1 first described from the Site 1039 reference section (Table T1, in PDF format). Thus, as expected, the nannofossil assemblages described from Site 1040 Unit U1 are nearly identical to those described for Site 1039 Unit U1.
From 371 to 404.81 mbsf, it is impossible to distinguish between latest Pleistocene to Holocene Zone NN21 from Zone NN20 because of the scarcity of all nannofossil species. Those species that are present are typically Pleistocene and extant, with no reworking present.
The top of P. lacunosa is observed at 405.74 mbsf and represents the top of early middle Pleistocene Zone NN19. The interval from 405.74 to 446.73 mbsf is assigned to Zone NN19. Nannofossils continue to be very scarce and poorly preserved in this interval. No reworking is apparent (Table T3, in PDF format).
D. brouweri is first observed at a depth of 449.8 mbsf. The LOD of D. brouweri defines the top of Zone NN18. Sediments from this level down to a depth of 459.4 mbsf are assigned to Zone NN18 (Table T3, in PDF format). Nannofossils in this interval are common to abundant. This indicates a change in lithology from the clayey diatomite of Unit U1, characteristic of the sediments between 371 and 422 mbsf, to the carbonate-bearing claystone of Unit U2 (Fig. F7).
Pliocene Zone NN17 is recognized in only one sample, at a depth of 462.4 mbsf, with the occurrence of the top of D. pentaradiatus (Table T3, in PDF format). Below this depth is a 10-m barren interval, which can only be assigned to the combined Pliocene Zones NN17-NN16. NN16 is defined by the LOD of Discoaster surculus to the LOD of D. pentaradiatus. Pliocene Zone NN15 is defined by the LOD of R. pseudoumbilica and/or the LOD of Sphenolithus. The tops of R. pseudoumbilica and Sphenolithus are observed at 472.75 mbsf, immediately below the 10-m barren interval. Although the LOD of D. surculus is not observed in the Site 1040 cores, if present, it should be observed in the 10-m barren interval.
The interval from 472.75 to 477.99 mbsf can only be assigned to undifferentiated Pliocene Zones NN15-NN12 because of the absence of several key FODs and LODs. Amaurolithus tricorniculatus, Discoaster asymmetricus, A. primus, Ceratolithus rugosus, Ceratolithus acutus, and Triquetrorhabdus rugosus, all key markers in defining the tops and/or bottoms of Pliocene Zones NN15, NN14, NN13, and NN12, are not observed in the Site 1040 cores (Table T3, in PDF format). Otherwise, nannofossils tend to be common to abundant in this interval. In the corresponding Site 1039 interval, Zones NN15-NN13 could not be differentiated because of the sporadic occurrence of the index species within that particular interval. At Site 1039, all of the marker species are scarce (Table T1, in PDF format).
The top of Discoaster berggrenii is observed at 477.99 mbsf and marks the top of Zone NN11, the top of the late Miocene, and at this site, the top of lithologic Unit U3. Lithologic Unit U3 extends to 653.53 mbsf and consists primarily of siliceous nannofossil chalk (Fig. F7). Nannofossils are generally abundant to highly abundant in this entire interval and are moderately to well preserved. Late Miocene Zone NN11 extends to 491.39 mbsf (Table T3, in PDF format).
The interval from 491.39 to 498 mbsf is assigned to late Miocene Zones NN10 through the top of NN9. The top of Coccolithus miopelagicus, present at 498 mbsf, corresponds to an LOD of 10.8 Ma, lying just above the base of Zone NN8. The top and bottom of Discoaster kugleri are located at 510 and 536.51 mbsf, respectively. The range between the FOD (11.8 Ma) and LOD (11.5 Ma) of D. kugleri may be used to approximate Zone NN7 in the absence of other markers. Thus, the interval between the top of D. kugleri and the top of C. miopelagicus (501-507 mbsf) can be assigned to either Zone NN8 or NN7. The interval from 510 to 536 mbsf is assigned to Zone NN7 (Table T3, in PDF format).
The difficulty in delineating Zones NN10-NN7 in these cores is due to the absence of Discoaster hamatus, D. surculus, Discoaster neorectus, and C. coalitus, which are all species used to define the tops or bottoms of these zones. In the corresponding Site 1039 lithologic interval, these species, while present, are very scarce with the exception of C. coalitus, which is not seen at Site 1039 either (Table T1, in PDF format).
The LOD of S. heteromorphus (13.6 Ma) defines the base of Zone NN6 and the top of Zone NN5. The top of S. heteromorphus is observed at 553.94 mbsf. This places Zone NN6 from 536.51 to 553.94 mbsf. Zone NN5 extends from 553.94 to 629.2 mbsf, the top of H. ampliaperta. The bottom of Zone NN5 is defined by the LOD of H. ampliaperta (15.6 Ma) (Table T3, in PDF format).
Zone NN4 is a gap zone defined by the LOD of Sphenolithus belemnos (18.2 Ma) to the LOD of H. ampliaperta. S. belemnos was not observed at Site 1040, nor was the FOD of H. ampliaperta or the FOD of S. heteromorphus. Thus, Zone NN4 extends from 629.2 mbsf to the bottom of the cored sedimentary sequence (653.52 mbsf) (Table T3, in PDF format).
Sediments in the bottom 9 m of Unit U3C of Hole 1040C are poorly preserved and not as diverse or abundant as seen upsection in lithologic Units U2B and U3, where assemblages are relatively well preserved, and very abundant to highly abundant. The bottom of Hole 1040C consists of 7.4 m of gabbro (Fig. F7).
The base of the sedimentary section just above the gabbro intrusion exhibits increased deformation, which shipboard structural geologists believe occurred shortly after sedimentation. If this is the case, then the gabbro cored here may represent basement that is between 15.6 and 18.2 Ma.
Site 1041 is located on the landward side of the Middle America Trench on the Caribbean plate at a water depth of ~3306 m, ~12 km landward (midslope) from the toe of the wedge and 10.8 km landward from Site 1040 (Figs. F1, F3, F4). A 500- to 600-m-thick sedimentary apron overlies the prism section at this location (Figs. F3, F7). Three holes were cored into the sedimentary apron, but the target depth of 550 m into the prism was not met. However, a total composite penetration of 423.8 mbsf was achieved.
A composite diagram of total core recovery, lithology, depth cored, and age of the three Site 1041 holes is illustrated in Figure F7. Hole 1041A penetrated 155.1 mbsf of apron sediments and recovered 18 cores. Total core recovery from Hole 1041A was 74.4%. Hole 1041B penetrated 395.6 mbsf and recovered 25 cores at a recovery rate of 55.4%. Hole 1041C was drilled to 395 mbsf before coring began. Three cores were taken from Hole 1041C, penetrating another 28.8 m beyond 395 mbsf. Total core recovery at Hole 1041C was 23.8%.
Shipboard sedimentologists have characterized the entire cored sequence below Core 170-1041A-2H as poor quality because of extensive drilling disturbance and large sections of rubble that strongly suggested considerable upper hole cavings (Kimura, Silver, Blum, et al., 1997). This, of course, has severe implications for all age dating attempts and is taken into account in the determination of the nannofossil biostratigraphy presented later in this study. Also in many sections, particularly from Cores 170-1041A-15X through 170-1041B-20R (350 mbsf), gas hydrate dissociation has obliterated sedimentary structures (Kimura, Silver, Blum, et al., 1997).
Only one lithologic unit comprises the entire interval drilled and cored at Site 1041 (Fig. F7). This unit is designated apron Unit A1. Unit A1 is divided into two smaller units: Subunit A1A is mostly clays and silts and Subunit A1B is coarser, made up mostly of silts and sands. Carbonate preservation is spotty in both units. Calcareous nannofossils are sufficiently abundant, however, to record a fairly good biostratigraphic record from the Site 1041 cores, despite widespread reworking of the sediments. Micritic calcite and dolomite rhombs are present in almost every observed sample from ~15 mbsf to the base of Unit A1, increasing in abundance downhole.
The distribution, abundance, and preservation of calcareous nannofossils recovered from Holes 1041A, 1041B, and 1041C correlated to the nannofossil zonation scheme of Martini (1971) are presented in Table T4, in PDF format. An age-depth plot is not possible because of the many limitations discussed below. However, a table of the range of ages with depth has been constructed from the nannofossil distribution data (Table T9).
There is little doubt that reworking of species occurs everywhere in the Site 1041 cores. An attempt has been made on the range-distribution chart to separate the reworked species from those deemed to be in situ to arrive at a meaningful biostratigraphy for Site 1041. Table T4, in PDF format summarizes this attempt; however, it is cautioned that all zones and all nannofossil datums observed are only relatively reliable. Conditions within this interval that contribute to the degree of reliability of the datums and zones are extensive reworking of older species, downhole contamination caused by cavings, and long intervals within the entire sequence that are barren of nannofossils, which severely limits the placement of observed tops and bottoms.
Nevertheless, taking into account all of the composite section's limitations to exacting an accurate determination of tops/bottoms, ages, and zonations at Site 1041, five ranges of nannofossil zones—or five divisions of time—are recognized. Referring to the Site 1041 nannofossil range chart in Table T4, in PDF format and the nannofossil zone definitions outlined in Table T8 will aid in understanding how each assigned range of nannofossil zones in this unit have been determined.
The interval from the top of the composite hole to a depth of 2 mbsf is assigned to combined nannofossil Zones NN21-NN19. This implies that the sediments may be <260 ka or as old as 1.95 Ma.
A relatively long interval, extending from 5 to 14.03 mbsf, is essentially barren of nannofossils but may be assigned to combined Zones NN21-NN17. The assignment restricts the age of this interval from <260 ka to as old as 2.55 Ma.
The interval from 14.03 to 134.25 mbsf is characteristic of nannofossil Zones NN18-NN16 relative to what is recognized both upsection and downsection. This interval may not be younger than 1.95 Ma or older than 3.75 Ma.
Samples in the interval from 137.17 to 275.21 mbsf may be assigned to combined nannofossil Zones NN16-NN11. This zonation represents a relatively longer length of time than the overlying intervals and indicates that this interval may be no younger than 3.75 Ma and no older than 8.6 Ma.
The interval from 280.7 to 415.54 mbsf can only be determined to be older than Zone NN11, or 8.6 Ma, which is the FOD of D. berggrenii.
Site 1042 is located at a water depth between 3593 and 3581 m on the Caribbean plate, landward of the Middle America Trench, ~7.5 km west of Site 1041 (Figs. F1, F4). Two holes were drilled and cored to a composite depth of 390.8 mbsf. Hole 1042A was spot cored; seven spot cores were taken over a depth of 240.1 mbsf (Fig. F7). Core recovery was 19.3%.
Hole 1042B was drilled to a depth of 316 mbsf and then cored to a depth of 390.8 mbsf. Eight cores were recovered from Hole 1042B with a recovery of 11.9% Two lithologic units have been described from the composite section. Unit 1 consists of silty claystone to 163 mbsf and silty claystone and limestone from 201.7 to 240 mbsf and from 353.7 to 390.8 mbsf (Fig. F7). The intervals from 163 to 201.7 mbsf and from 240 to 316 mbsf were not cored.
Unit U2, from 316 to 342.9 mbsf, consists of a carbonate-cemented sandstone breccia above a multilithic breccia with a clayey-silt matrix present at 342.9-353.7 mbsf. From 353.7 mbsf to the base of the hole at 390.8 mbsf is a silty claystone that is interpreted as Unit I.
Unfortunately, the top 250 mbsf at Site 1042 was spot cored every 50 m, so a complete biostratigraphy for that section is not obtainable. Seven rotary cores were recovered. Several wiper trips and hole sweeps were conducted while we attempted to recover Cores 170-1042A-6R and 7R, adding downhole contamination to an already overworked hole (Kimura, Silver, Blum, et al., 1997). Because of the considerable amount of reworking, downhole contamination, and poor core recovery, no age determination or reliable biostratigraphy is possible utilizing calcareous nannofossils. In fact, neither diatoms nor foraminifers are able to provide further information about the upper 250 mbsf.
A second hole (Hole 1040B) was rotary-drilled to ~313 mbsf and then cored to 383 mbsf (Fig. F7). Calcareous nannofossil age determinations for this interval can not be determined with any degree of confidence.
During shipboard analysis, there was a somewhat optimistic attempt to assign nannofossil zones to the Site 1042 sequence (see p. 202, table 6, in Kimura, Silver, Blum, et al. [1997]). Although several zones were tentatively recognized, they are far from conclusive and should not be considered accurate. That nannofossil distribution and abundance range chart (Table T5) is reproduced here without the zones assigned aboard ship. In spite of the minimal biostratigraphy accomplished, the nannofossil distribution does give some sense of superposition at Site 1042.
Site 1043 (water depth = 4321 m) is located 0.4 km northeast of the toe of the continental slope on the Caribbean plate and 2.06 km from Site 1039 on the Cocos plate across the Middle America Trench, along seismic line CR-20 (Figs. F1, F2, F4). Site 1043 was cored through the sedimentary accretionary wedge and the décollement and into the underthrust section initially described from the Site 1039 reference cores taken from the immediate seaward side of the trench on the Cocos plate. Only one hole was cored at Site 1043. Hole 1043A was continuously cored to a depth of 282.3 mbsf. Total core recovery was 79.6%.
Only one lithologic unit is defined from the wedge sediments from Hole 1043A (Fig. F7). Lithologic Unit T1 is composed of thick intervals of silty clay and clay extending from the seafloor to 150.7 mbsf. These sediments are very characteristic of poorly sorted submarine debris and finer-grained sediment flows. Scattered throughout the sediments are clasts of clay, nannofossil ooze, and siliceous clayey nannofossil ooze. Overall, nannofossil abundances range from few to abundant within this entire unit. The base of unit T1 is the décollement.
The décollement is defined in Hole 1043A as a zone of anastomosing, discontinuous polished fracture systems observed from Section 170-1043A-16X-2, 112 cm, to the bottom of 17X-2 (141.50-150.57 mbsf), although the base is not sharply defined (Kimura, Silver, Blum, et al., 1997) (Fig. F7). Diatom abundances increase from trace amounts within the décollement zone to common immediately below the base of the zone, between 149.36 and 157.55 mbsf (Kimura, Silver, Blum, et al., 1997).
Below the décollement in the underthrust sediments, three lithologic units are recognized (Fig. F7). Lithologic Unit U1 extends from 150.57 (a structurally placed base) to 193.95 mbsf. The 5.5-m-thick turbidite observed in Unit U1 at Site 1039 is not observed here. Here, Unit U1 is predominately a diatomaceous ooze with varying amounts of clay and carbonate (Kimura, Silver, Blum, et al., 1997). Nannofossils are common in this sequence.
Lithologic Unit U2, extending from 193.95 to 263.71 mbsf, is characterized by a marked decrease in biogenic sediments and an increase in clay-sized terrigenous sediments. The tops of Unit U2 at both Sites 1039 and 1043 exhibit the same changes in the biogenic and terrigenous components. Unit U2 was divided into Subunits U2A and U2B based on the presence of strata containing siliceous nannofossil ooze and calcareous clay, with nannofossils in the latter (Kimura, Silver, Blum, et al., 1997). Subunit U2B extends from 245.63 to 263.71 mbsf. In general, there is an increase in nannofossil abundance from the top of Unit U2 to its base (Fig. F7).
Lithologic Unit U3 is composed of siliceous nannofossil ooze and extends from 263.71 to 282.3 mbsf. Unit U3 was subdivided at Site 1039 into Subunits U3A, U3B, and U3C based on variations in the biogenic component downcore. Because of the depth cored, only Subunit U3A is observed at Site 1043. The bottom of Subunit U3A was not reached at this site (Fig. F7). Nannofossils are very abundant with depth throughout most of this interval until 281.65 mbsf, where their abundance drops by approximately an order of magnitude.
The distribution, abundance, and preservation of calcareous nannofossils recovered from Site 1043 correlated to the nannofossil zonation scheme of Martini (1971) is presented in Table T6, in PDF format. An age-depth diagram plotting the age of all observed nannofossil datums at Site 1043 with depth is shown in Figure F10.
The entire Unit T1 is Pleistocene in age, based on the consistent observations of Gephyrocapsa oceanica, Gephyrocapsa caribbeanica, H. sellii, P. lacunosa, and Helicosphaera neogranulata in the cored sequence from the seafloor down to the base of the décollement at 150.57 mbsf. Nannofossil Zones NN21, NN20, and NN19 are recognized in Unit T1 (Table T6, in PDF format).
Nannofossil Zone NN21 extends from the seafloor to 7.99 mbsf, the observed top of E. huxleyi (FOD = 260 ka). Middle and late Pliocene reworked species are observed. Although less abundant, the number of reworked species seen in this interval equals the number of extant species observed.
From 7.99 to 28.2 mbsf, Zone NN20 is recognized by the absence of E. huxleyi and P. lacunosa. Preservation is very poor in this interval, and individual specimens are scarce. However, there appears to be enough of an assemblage to assign this interval a zone, despite moderate reworking of middle and late Miocene calcareous nannofossil species.
The top of P. lacunosa is observed at 29.76 mbsf. The LOD of P. lacunosa defines the base of Zone NN20 and the top of Zone NN19. Zone NN19 extends from 29.76 to 150.57 mbsf (base of décollement).
The base of Zone NN19 is defined by the LOD of D. brouwerii, which was not seen in Unit T1. Therefore, with reasonable certainty, Unit T1 appears to be entirely Pleistocene.
Carbonate clasts within Unit T1 were also analyzed for the presence of nannofossils. One clast at 92.05 mbsf contains a late Miocene Zone NN11 assemblage. Another clast, sampled at 129.66 mbsf, is typically early to middle Miocene Zone NN4. The nannofossil assemblage observed in this clast is identical to the assemblages observed in the oldest sediments cored at Sites 1039 and 1040 (Tables T1, in PDF format; T3, in PDF format).
The décollement extends from 140.5 to 150.57 mbsf. Nannofossils, although present and indicative of Zone NN4, are not common within the décollement zone.
Five nannofossil zones are recognized in Subunit U2A, which extends from 150.57 to 245.63 mbsf. From 150.57 to 171.8 mbsf, late Pleistocene-Holocene Zone NN21 is observed based on the occurrence of E. huxleyi at 171.8 mbsf. Two of the five samples taken at regular intervals in this zone are barren, and the preservation is generally poor to moderate (Table T6, in PDF format).
Nannofossil Zone NN20, established by the absence of both E. huxleyi and P. lacunosa, is assigned to the interval from 174.8 to 181.6 mbsf. Below this level, Zone NN19 is recognized by observing both the top of P. lacunosa at 181.89 mbsf and the top of D. brouwerii at 226.5 mbsf (Table T6, in PDF format).
Zone NN18, defined by the LOD of D. pentaradiatus to the LOD of D. brouwerii, cannot be delineated in the interval from 226.5 to 241.9 mbsf because the LOD of D. pentaradiatus is not observed at this site. The LOD of D. surculus is not observed either; this LOD defines the top of Zone NN16 and base of Zone NN17. R. pseudoumbilica, whose LOD defines the bottom of Zone NN16, is observed at 245.67 mbsf. Thus, the interval from 226.65 to 242.9 mbsf is assigned to undifferentiated Zones NN18-NN16. Nannofossils in this strata are typically few in number and poorly preserved. The base of Subunit U2A approximates the base of the strata assigned to Zone NN18-NN16 (Table T6, in PDF format).
At Sites 1039 and 1040, some difficulty exists in delineating Zones NN18-NN16 and NN15-NN12, a situation similar to that at Site 1043. The periods of time represented by these assemblages exhibit the lowest sedimentation rates (see Figs. F8, F9, F10).
Nannofossils in Subunit U2B are abundant and generally better preserved than those upsection in Subunit U2A and show identical assemblage characteristic trends observed in the Sites 1040 and 1039 underthrust sequences. Despite increased preservation and abundance, Zones NN15-NN12 cannot be resolved between 251.68 and 258.3 mbsf. Samples from that interval are assigned to undifferentiated Zones NN15-NN12. Recall that at Sites 1039 and 1040 in the underthrust sections the marker species in the Zones NN18-NN12 interval are scarce or not present. These include A. tricorniculatus, A. primus, C. rugosus, D. asymmetricus, and T. rugosus (Table T6, in PDF format).
Nannofossils in Unit U3 are very to highly abundant and diverse and generally exhibit good preservation (Unit U3 is a siliceous calcareous nannofossil ooze). At Site 1043, all of Unit U3 is placed in Zone NN11, which is defined as the FOD to LOD of Discoaster quinqueramus. Unlike the lithology cored at Sites 1039 and 1040, the base of Unit U3 was not reached at Site 1043.