SYNTHESIS AND RESULTS

Lithostratigraphy

Sediments recovered during Leg 182 record carbonate deposition in a mid- and high latitude setting against the background of an evolving Southern Ocean and northward drift of the Australian continent. Sediment in 3875 m of water at the toe of slope (Site 1128) chronicles the change from early Paleogene time, when a humid onshore climate flushed large amounts of terrigenous clastic sediment into the deep sea, to Neogene time, when increasing continental aridity promoted mostly marine carbonate deposition. Green Eocene siliciclastic sands and silts that accumulated largely below the carbonate compensation depth (CCD) in a poorly oxygenated, bathyal setting become finer grained upward, with much of the deep-water late Eocene represented by clay deposition (Fig. 3). Initiation of the contemporary Southern Ocean circulation system, and thus the modern global ocean, is signaled by a gradual change to lower Oligocene brown clay and carbonate, as this part of the seafloor became ventilated and the CCD deepened. The deep-water Neogene record is one of pink pelagic carbonate ooze punctuated by white planktonic foraminiferal turbidites. Early-middle Miocene time is represented by a major hiatus and sediment gravity-flow deposits.

Most drilling took place on the upper slope and outermost shelf, in 202-784 m of water, through a mainly carbonate succession (Figs. 3, 4). Two distinct groups of strata, Eocene-middle Miocene and late Miocene-Quaternary in age, form the upper part of the continental margin. The older succession is stratigraphically equivalent to and roughly coeval with most of the Eucla Group, exposed onshore in the Eucla Basin beneath the Nullarbor Plain. Sediments are a package of Eocene shallow-water terrigenous sands and carbonates that deepens upward into Oligocene and early-middle Miocene pelagic ooze and chalk (Fig. 3). Although recovery was generally poor through the Oligocene and Miocene interval because of silicification (Site 1134 is an exception for the Miocene), recovered carbonates are characterized by stained hardgrounds and numerous omission surfaces. The younger, wholly Neogene succession is a large, seaward-dipping wedge of carbonate sediment that downlaps onto the older sediments and has been prograding seaward onto the Eyre Terrace since late Miocene time. The contact between the two successions is represented, particularly the late Miocene and especially the Pliocene, by slumps, sediment gravity-flow deposits, or unconformities. Such erosion, corrosion, and/or mass-wasting and redeposition processes reflect periods of margin instability, seismicity, or lowered sea level.

The most astonishing discovery of Leg 182 was that the huge wedge of slope sediment prograding onto the Eyre Terrace is nearly entirely Pleistocene in age (Figs. 3, 4). This deposit, formed by carbonate produced on the outer shelf and upper slope and swept seaward, is more than 500 m thick in the center (Fig. 4), where rates of accumulation exceed 40 cm/k.y., equivalent to many shallow-water tropical carbonates and twice the rate of Bahamian slope sedimentation. The green and gray material is surprisingly uniform in composition, made up of fine carbonate sand and silt that was reworked in place by generations of burrowing organisms, leading to multitiered trace fossil assemblages. Particles are all skeletal fragments, mainly delicate bryozoans, ostracodes, benthic and planktonic foraminifer tests, tunicate sclerites, nannofossils, and siliceous sponge spicules. The deposit grades from mud and sand at the top downward to partially lithified sediment in the middle, and finally to hard limestone at the base. Sediment below 400 mbsf is usually neomorphosed with numerous fossil molds, or partially altered to sucrosic dolomite.

The facies transition upslope into shallower water is marked by the presence of numerous bryozoan-rich buildups (Figs. 3, 4). These mounds, in water depths of ~200-350 m, are dominantly muddy and characterized by the prolific growth of numerous and diverse bryozoans. The mounds in particular have seafloor relief of as much as 20 m and extend laterally many hundreds of meters. These are among the first modern analogs to similar mounds that were an important part of the carbonate depositional systems in earlier, Phanerozoic time.

Biostratigraphy

Age of Sediments and Hiatuses

The Cenozoic sequence penetrated at Leg 182 sites mainly represents the Pleistocene, Miocene, Oligocene, and Eocene. Biostratigraphic data indicate that these Cenozoic successions are mostly hiatus bound (Fig. 5). The Pliocene and the later part of the middle Miocene successions are either missing or highly condensed. Other intervals of the Miocene and Oligocene are also missing, though unmatched between sites along the eastern and western transects.

A greatly expanded Pleistocene section was recovered at shallow-water sites from the eastern transect (Sites 1129, 1131, and 1127). The longest at Site 1129 exceeds 540 mbsf, with an average accumulation rate of more than 300 m/m.y. (Fig. 6). In contrast, the Pleistocene sections from the western transect only extend down to ~250 mbsf at Sites 1130, and ~230 mbsf at Site 1132. The biostratigraphic units, however, are similar between eastern and western transect sites. The Pleistocene section is underlain by a thin upper Miocene interval, which in turn disconformably overlies a unit of mainly early-middle Miocene age. The absence of many Pliocene to later middle Miocene biozones from Sites 1127, 1129, and 1130 signifies a hiatus of at least 7 m.y. between the Pleistocene-Pliocene and the middle Miocene. This major unconformity was detected at all other sites as two shorter hiatuses of ~3 m.y. each at the middle/late Miocene and late Miocene/Pliocene-Pleistocene boundaries. They are more clearly recorded as disconformities capping the upper Miocene section at Sites 1126, 1134, and 1133 in intermediate water depths, and at shallow-water Sites 1130 and 1132 along the western transect (Fig. 5). Thus, it is reasonable to speculate that the unconformity at the Pleistocene-Pliocene/middle Miocene boundary, as observed at eastern sites, probably resulted from at least two major events that subsequently almost completely erased the sedimentation record of the later middle Miocene-Pliocene.

The late Miocene succession is better represented in the west, especially at Site 1130 and at deep-water sites. The middle Miocene, mainly the early part, is present at all sites except Sites 1128 and 1130. It also represents the oldest sediment penetrated at most sites along the eastern transect, although a thin lower Miocene unit was encountered at the base of Hole 1131A (~590 610 mbsf). An expanded middle Miocene section at Site 1132 was accompanied by sedimentation rates ranging 14 54 m/m.y. (Fig. 6). A similar rate of 50 m/m.y. was recorded for the middle Miocene at Site 1133. The absence of middle-early Miocene biofacies from Sites 1130 and 1128 is interesting because these two sites respectively represent the intermediate and deepest water sites along the western transect. At Site 1130 (495 mbsf), the upper Miocene section was disconformably underlain by sediments of mainly early-middle Oligocene age, indicating a depositional gap of ~15 m.y. At Site 1128 (3890 mbsf), the entire middle Miocene-upper Oligocene interval corresponds to a ~15-m-thick debrite with mixed calcareous nannofossil and planktonic foraminifer assemblages of early to late Miocene age, suggesting unconformities totaling ~13 m.y. in duration. Sites 1126 and 1134, the two intermediate-water sites from the west, exhibit remarkable similarities in assemblage compositions, ages of recovered sediments, and position and number of hiatuses. These are the only sites yielding definite early Miocene calcareous nannofossils and planktonic and benthic foraminifers.

Sediments of Oligocene age and older were recovered only from the west. At Sites 1130 and 1132, the Oligocene section disconformably overlies Eocene sediments, and a hiatus of at least 2 m.y. is indicated by biostratigraphic data. However, the zonal succession of nannofossils and planktonic foraminifers is largely continuous across the Oligocene/Eocene boundary, suggesting a conformable succession at Sites 1126, 1128, and 1134. Mainly early Oligocene assemblages are present at Sites 1132, 1130, and 1128 in the west, and an expanded lower Oligocene section was indicated by nannofossils at Site 1128. Biostratigraphic resolution below 70 mbsf at Site 1128 was largely achieved by calcareous nannofossils because planktonic foraminifers were rare or barren in sediments dominated by siliceous oozes and packstones.

Holes 1126D and 1134A are the only holes containing rich calcareous microfossils of Eocene age. Poor preservation, however, impaired proper recognition of species and biozones in some intervals. Although present at all western sites, the middle Eocene is represented by poorly preserved, impoverished assemblages in various poorly recovered lithologies. It consists of a siltstone (~150 m thick) at Site 1128 with sporadic nannofossils, a calcareous packstone (~30 m thick) at Site 1126 with moderately preserved nannofossils and planktonic and benthic foraminifers, and a dark, iron-stained sand (30 m thick) at Site 1134 with rare and poorly preserved microfossil assemblages. A calcareous sandstone at Site 1130 and a bioclastic limestone at Site 1132 both contain shallow-water associations of planktonic and benthic foraminifers. No calcareous nannofossils or foraminifers were discovered in dark green sandstone at the base of Hole 1126D, which may correspond to Cretaceous synrift sediments along the southern Australian margin.

Paleobathymetry and Paleoceanography

The succession of microfossil assemblages from Leg 182 provide an unmatched record of sea-level and circulation changes in the Great Australian Bight from the Eocene to Holocene. Expanded Pleistocene successions at Sites 1127, 1129, 1130, 1131, and 1132 contain upper bathyal benthic foraminifer assemblages with a redeposited neritic component. Apparent fluctuations in the relative proportions of neritic taxa and upper bathyal taxa indicate that downslope transport periodically varied in intensity during the Pleistocene. These changes, together with the periodic intrusion of warm-water species into temperate planktonic foraminifer assemblages, indicate responses to changing climate and oceanic circulation. Calcareous nannofossil assemblages dominated by Braarudosphaera bigelowii indicate that a major ecological crisis affected the planktonic ecosystem at base of the Pleistocene.

At the three shallow-water sites (Site 1129, 1131, and 1132), diversified, extremely well preserved benthic foraminiferal assemblages, including unusually large specimens (>1 mm), are found together with abundant and well-preserved bryozoan fragments (Fig. 5). These assemblages were probably part of a highly dynamic ecosystem that became established at the seafloor during the later Pleistocene, coincident with bryozoan buildups in paleodepths of 200-300 m. A shallowing-upward trend from middle bathyal paleodepths in the middle and late Miocene to upper bathyal paleodepths in the Pleistocene is recorded at Sites 1129, 1130, and 1132 (Fig. 5). Middle bathyal faunas at these sites reflect middle and late Miocene transgressive pulses that are also expressed onshore southern Australia by thin transgressive tracts.

Middle bathyal assemblages from upper Eocene-upper Miocene successions show major changes in composition at intermediate water Sites 1126 and 1134. The most severe changes in the early Miocene probably relate to global changes in deep- and intermediate-water circulation affecting the distribution of bathyal benthic foraminifers worldwide. The composition of abyssal assemblages at Site 1128 indicates deposition above the CCD during the Pleistocene, close to lysocline in the late Oligocene-middle Eocene, and below the CCD in early-middle Eocene in the deep basin (Fig. 5). Assemblage boundaries appear to be coeval in the Eocene, Oligocene, and lower Miocene of intermediate- and deep-water Sites 1126, 1128, and 1134, suggesting that benthic foraminifer distribution was controlled by major environmental changes during these periods. Discrepancies between assemblage boundaries of deep-, intermediate-, and shallow-water sites in the middle Miocene-Pleistocene sequence relate to stratigraphic or lithological differences, and reflect distinct depositional and paleoceanographic regimes in the region.

Paleomagnetics

Previous carbonate legs have proved very unsatisfactory for paleomagnetists because the magnetic intensity of the sediments has been so low that remanence measurements were essentially unreliable. However, with the change to direct-current superconducting quantum interface devices (DC SQUIDs) in the new instrument, it was possible to measure the magnetization at most sites during Leg 182. Nevertheless, the natural remanent magnetization (NRM) of sediments from Sites 1126 and 1134 was so weak that after a coring overprint was removed the magnetization was at the noise level of the instrument for much of the hole.

Coring Experiments

During Leg 182, an experimental nonmagnetic cutting shoe was used in advanced hydraulic piston core (APC) coring. The shoe and other components of the bottom-hole assembly were used to continue investigations of anomalous magnetization observed in these cores. Comparisons were made between cores taken with standard core-barrel assemblies, with standard core-barrel assemblies and the nonmagnetic shoe, and with nonmagnetic core barrels and shoe. On some occasions, it was clear that the nonmagnetic shoe and assembly greatly improved the paleomagnetic record. On other occasions, the shoe alone improved the record compared with the core-barrel standard assemblies. However, there were also sites where neither the nonmagnetic shoe nor the whole nonmagnetic core-barrel assembly had any obvious effect. This is consistent with the intermittent nature of the anomalous magnetization experienced during past legs. Work postcruise will focus on trying to isolate other sediment properties such as grain size, porosity, or shear strength, which may affect the coring contamination.

Rock Magnetism

A common feature of the rock magnetism of the upper sections of Leg 182 holes was the extremely high ratio of anhysteretic remanent magnetization to isothermal remanent magnetization (ARM/IRM). This is a measure of the degree to which single-domain particles dominate among the magnetic phases present in the samples. This ratio generally decreased downhole and was accompanied by a decrease in intensity of magnetization and coercivity. This decrease in intensity of magnetization is generally interpreted as dissolution of the finest grained magnetic phases in the sediment. IRM acquisition suggests that magnetite and magnetic sulfides are the principal remanence carriers. The observation of a downhole decrease in ARM/IRM and intensity is consistent with diagenesis models of organic matter oxidation and sulfate reduction as the principal processes that regulate the preservation of ferromagnetic phases.

Magnetostratigraphy

Long-core and discrete sample measurements yielded a record of inclination from which a Pliocene-Pleistocene magnetostratigraphy was interpreted. Overall, deeper water records are of good to fair quality, yielding accurate estimates of the expected inclination for the present latitude. In contrast, lower intensities in the shallow-water sites result in shallow inclinations with large scatter. The Brunhes/Matuyama boundary was identified in five high-sedimentation-rate sites (Fig. 7; Table 1). The Jaramillo Subchron (C1r1n) and the top of the Olduvai Chron (C2n) were also found at three sites. The interpretation of magnetic polarity in Figure 7 implies that sedimentation rates in the early Pleistocene were considerably lower than during the Brunhes Chron. In addition, variations of the intensity of the remanence after partial demagnetization were found to oscillate on time scales comparable with those of the geocentric axial dipole. Although the intensities have not been normalized to account for variations in concentration, these sediments may provide a valuable relative paleointensity record that can be used for high-precision correlation between sites. For Sites 1126 and 1134, magnetostratigraphic data were obtained for parts of the Miocene. Site 1128 also contains a long record of the polarity sequence of the early Oligocene.

Geochemistry

The drilling of nine sites off the Eucla margin provided a unique insight into the fluid dynamics along a continental margin dominated by cool-water carbonates. With respect to this objective, the most significant discovery was the presence of a brine, varying in salinity between 80 and 105, that was present in and underlying seven of the nine sites (the exceptions being Site 1128, drilled in a water depth of 3884 m, and Site 1133). At Site 1133 (located in 1043 m of water) fluids were encountered with a salinity of 40, but recovery problems precluded the measurement of the salinities in deeper samples. The brine was present at relatively shallow depths at the deeper sites (1134 and 1126), whereas the maximum salinities were not encountered until depths ≥400 mbsf at the shallower sites (1127, 1129, 1130, 1131, and 1132). The impact of the brine is more readily visible along the eastern transect (Sites 1129, 1131, and 1127; see also Fig. 4). Similar processes appear to occur at Sites 1130 and 1132, although they are reduced in nature.

At Sites 1129, 1131, and 1127, a gradual increase in salinity was encountered, reaching a maximum value of ~100 at Site 1127 (farthest away from the margin) and 92 at Site 1129 (closest to the platform). A contour map of the distribution of Cl- at the three sites is shown in Figure 8. Pore fluids in the Pleistocene portion of the sediments from these sites also possess a Na+/Cl- ratio exceeding that of seawater, suggesting that the fluids in the sediments were involved in the dissolution of NaCl. Although the origin of the fluids has not yet been established, a probable explanation is that they formed during the Pleistocene when the Eucla shelf was exposed numerous times during sea-level lowstands. We suggest that large hypsersaline lagoons developed during these episodes of low sea level. Because of the greater hydrostatic head, fluids with high salinity were forced into underlying strata and out onto the adjacent continental slope. During sea level highstands high salinity fluids then diffused upward as additional sediment was deposited, and the profiles of the individual ions were modified by diagenetic reactions in the sediments. This sequence may have occurred a number of times as sea level fluctuated during the Pleistocene. This hypothesis is supported by an examination of the Cl- distribution between the three sites (Fig. 8) which suggests that the top of the brine has a common depth below sea level and, therefore, crosscuts sequence boundaries.

All three sites exhibited high concentrations of hydrogen sulfide (H2S) and methane, combined with high values of alkalinity. The high concentrations of H2S are derived from the oxidation of organic material by sulfate-reducing bacteria. The relatively low concentrations of iron in carbonates, in contrast to those found in locations adjacent to continental margins with a large amount of siliciclastic input, means that the H2S is not sequestered as iron sulfides. Consequently, concentrations of H2S are able to reach high levels (>150,000 ppm at Site 1129). As a result of the high rate of sedimentation (>200 m/m.y.) and the position of the sites close to the continental shelf, these sites contained an initial high concentration of organic material. Under normal conditions, the organic material would be oxidized first by oxygen and then by sulfate-utilizing bacteria, thereby creating alkalinity and hydrogen sulfide. Although this process also takes place in the Eucla margin, the high-salinity brines underlying and within the Pleistocene succession provide as much as three times the normal sulfate concentrations; therefore, with sufficient organic material, significantly higher amounts of hydrogen sulfide can be formed. As the sediments are buried, the concentration of sulfate is depleted in the sediment and a gradient is established between the sediments, the overlying seawater, and the underlying brine. Because of the high sulfate concentration of the brine, the flux of sulfate diffusing into the zone of organic material remineralization is significantly greater than in a normal marine sediment, thus resulting in greater than normal sulfate reduction and higher alkalinity. In addition to H2S, high concentrations of methane and the presence of other higher molecular-weight hydrocarbons were discovered in the sediments at Sites 1129, 1131, and 1127. At Site 1127, gas pockets in the core were found to contain as much as 73% methane and 365 ppm ethane. We suggest that these gases, including the higher molecular-weight hydrocarbons, are bacterial in origin, being generated in situ. Higher molecular-weight alkanes are generally only believed to form through thermogenic processes (Fig. 9). However, the concentration of ethane, propane, and butane did not increase with depth, but instead reached a maximum associated with the maximum in methane, suggesting that these gases also originated in this portion of the sediment and were not thermogenic in origin.

The oxidation of organic material also has an important influence on the process of carbonate recrystallization. The sediments at all the sites cored during Leg 182 initially contained a mixture of aragonite, high-Mg calcite (HMC), and low-Mg calcite (LMC). Of these minerals, aragonite and HMC are metastable and gradually will alter to the more stable forms of LMC and dolomite. This process is evident at all the sites drilled and is greatly accelerated by the processes of sulfate reduction and the consequent formation of H2S. During the oxidation of organic material, two important chemical reactions take place that alter the carbonate chemistry in the pore waters: (1) sulfate reduction creates two moles of alkalinity for every mole of sulfate that is consumed and (2) one mole of H2S is produced (2CH2O + SO42- = H2S + 2HCO3- ). The H2S dissociates in the pore water causing the pH to be reduced and the speciation of carbonate in the pore waters to move from being dominated by HCO3- to H2CO3. This sequence of events produces pore fluids that are undersaturated with respect to the metastable carbonate minerals. The most thermodynamically soluble form of calcium carbonate, HMC dissolves first, followed by aragonite, and LMC. This trend is followed at all of the sites. The high-alkalinity environment creates a thermodynamic regime favorable for the formation of dolomite. Dolomitization consumes Mg2+ from the pore waters, setting up a strong diffusive gradient into the reaction zone from the overlying seawater and the underlying brine. The formation of dolomite is well illustrated at Sites 1127, 1131, 1129 (Fig. 10), which shows concentrations of ~5% at Sites 1127 and 20% at Sites 1129 and 1131.

Although the high-salinity pore fluids were also present at other sites (Sites 1130, 1132, 1126, and 1134), these sites showed lower concentrations of H2S and methane and consequently lower amounts of carbonate diagenesis. The controlling factor at these sites is the supply of organic material combined with the rate of sedimentation.

Petrophysical and Downhole Measurement Data

Downhole logging data were collected at eight of nine sites drilled during Leg 182. To complement these data, an extensive program of sediment physical properties measurements was undertaken. Both datasets will provide detailed information for intersite correlation, characterization of lithostratigraphic boundaries, and core-log correlation. The presence of chert, particularly in the post-Pleistocene sequences sedimentary section, resulted in low recovery. Downhole logging data will provide the information necessary to fill in recovery gaps, refine the placement of lithostratigraphic boundaries, and enable interpretation of sedimentary facies, composition, and structure within missing intervals. Checkshot surveys conducted at five sites will facilitate correlation of the drilled strata to the regional grid of high-resolution seismic data to formulate and refine models of temperate-water carbonate sedimentation and view the influence of eustatic sea level variations on this sedimentation.

The downhole logging and physical properties measurement program undertaken during Leg 182 provided an unparalleled opportunity to investigate early and postdepositional diagenesis in a nontropical carbonate depositional system. Preliminary diagenetic characterization of temperate-water carbonate sediments recovered during Leg 182 show distinct geochemical, mineralogic, and textural variations, despite the preconceived notion that cool-water carbonates have low diagenetic potentials because of their dominantly low-magnesium calcite mineralogy. Postcruise analysis of Formation MicroScanner (FMS) and conventional logs will provide a more detailed analysis of mineralogical and textural changes occurring during diagenetic alteration.

An important finding of Leg 182 was the discovery of distinct, possibly Milankovitch, cyclicity recorded in gamma-ray and density data from the thick Pleistocene sediment wedge drilled at numerous sites (Fig. 11). The high sedimentation rates of this deposit, in excess of 40 cm/k.y., will enable frequency analysis of the logging and physical properties datasets to assist in the refinement of the biostratigraphic ages within this time interval. FMS data collected at six of the logged holes will allow for detailed investigation of depositional facies and sedimentary structure in the sediment cycles, particularly in the intervals where recovery was poor. Correlation of the cyclicity present in the temperate-water carbonate wedge to other Pleistocene carbonate and noncarbonate marginal systems will provide important information on differences in eustatically influenced sedimentation patterns occurring on continental margins from the seafloor to the bottom of the drilled interval.

To 182 Principal Results Site 1126

To 182 Table of Contents