Continuous Flow Analysis Method for Determination of Dissolved Reactive Phosphorus in Ice Cores

Phosphorus (P) is an essential macronutrient for all living organisms. Phosphorus is often present in nature as the soluble phosphate ion PO43– and has biological, terrestrial, and marine emission sources. Thus PO43– detected in ice cores has the potential to be an important tracer for biological activity in the past. In this study a continuous and highly sensitive absorption method for detection of dissolved reactive phosphorus (DRP) in ice cores has been developed using a molybdate reagent and a 2-m liquid waveguide capillary cell (LWCC). DRP is the soluble form of the nutrient phosphorus, which reacts with molybdate. The method was optimized to meet the low concentrations of DRP in Greenland ice, with a depth resolution of approximately 2 cm and an analytical uncertainty of 1.1 nM (0.1 ppb) PO43–. The method has been applied to segments of a shallow firn core from Northeast Greenland, indicating a mean concentration level of 2.74 nM (0.26 ppb) PO43– for the period 1930–2005 with a standard deviation of 1.37 nM (0.13 ppb) PO43– and values reaching as high as 10.52 nM (1 ppb) PO43–. Similar levels were detected for the period 1771–1823. Based on impurity abundances, dust and biogenic particles were found to be the most likely sources of DRP deposited in Northeast Greenland.

New Data on Welshite, e.g. Ca2Mg3.8Mn0.62+Fe0.12+Sb1.55+O2 [Si2.8Be1.7Fe0.653+Al0.7As0.17O18], an Aenigmatite-Group Mineral

Electron and ion microprobe data on two samples of welshite from the type locality of Langban, Sweden, gave analytical totals of 99.38-99.57 wt.% and BeO contents of 4.82-5.11 wt.%, corresponding to 1.692-1.773 Be/20 O. Mossbauer and optical spectra of one of these samples gave Fe-[iv](3+)/Sigma Fe = 0.91, Fe-[iv](2+)/Sigma Fe = 0.09, and no evidence of Mn3+. The resulting formula for this sample is Ca2Mg3.8Mn0.62+Fe0.12+Sb1.55+O2[Si2.8Be1.7Fe0.653+Al0.7As0.17O18], and that for the second sample, Ca2Mg3.8Mn0.12+Fe0.12+F0.83+Sb1.25+O2[Si2.8Be1.8F0.653+Al0.25As0.25O18], is related by the substitution involving tetrahedral and octahedral sites: 0.59([vi,iv])(Fe,Al)(3+) approximate to 0.42([vi])(Mg,Mn,Fe)(2+) + 0.21(Sb-[vi],As-[iv])(5+), i.e. 3([vi,iv]) M3+ = 2([vi])M(2+) + M-[vi,iv](5+). WelShite is distinctive among aenigmatite-group minerals in the high proportion of Fe 3+ in tetrahedral coordination and is unique in its Be content, substantially exceeding 1Be per formula unit. Given the cation distributions in other minerals related to aenigmatite, we think it is reasonable to assume that at least one tetrahedral site is >50% occupied by Be and that one octahedral site is >50% occupied by Sb, so that welshite should be retained as a distinct species with its own name in the aenigmatite group.

Hyalotekite, (Ba,Pb,K)(4)(Ca,Y)(2)Si-8(B,Be)(2)(Si,B)(2)O28F, a Tectosilicate Related to Scapolite: New Structure Refinement, Phase Transitions and a Short-Range Ordered 3B Superstructure

Hyalotekite, a framework silicate of composition (Ba,Pb,K)(4)(Ca,Y)(2)Si-8(B,Be)(2) (Si,B)(2)O28F, is found in relatively high-temperature(greater than or equal to 500 degrees C) Mn skarns at Langban, Sweden, and peralkaline pegmatites at Dara-i-Pioz, Tajikistan. A new paragenesis at Dara-i-Pioz is pegmatite consisting of the Ba borosilicates leucosphenite and tienshanite, as well as caesium kupletskite, aegirine, pyrochlore, microcline and quartz. Hyalotekite has been partially replaced by barylite and danburite. This hyalotekite contains 1.29-1.78 wt.% Y2O3, equivalent to 0.172-0.238 Y pfu or 8-11% Y on the Ca site; its Pb/(Pb+Ba) ratio ranges 0.36-0.44. Electron microprobe F contents of Langban and Dara-i-Pioz hyalotekite range 1.04-1.45 wt.%, consistent with full occupancy of the F site. A new refinement of the structure factor data used in the original structural determination of a Langban hyalotekite resulted in a structural formula, (Pb1.96Ba1.86K0.18)Ca-2(B1.76Be0.24)(Si1.56B0.44)Si8O28F, consistent with chemical data and all cations with positive-definite thermal parameters, although with a slight excess of positive charge (+57.14 as opposed to the ideal +57.00). An unusual feature of the hyalotekite framework is that 4 of 28 oxygens are non-bridging; by merging these 4 oxygens into two, the framework topology of scapolite is obtained. The triclinic symmetry of hyalotekite observed at room temperature is obtained from a hypothetical tetragonal parent structure via a sequence of displacive phase transitions. Some of these transitions are associated with cation ordering, either Pb-Ba ordering in the large cation sites, or B-Be and Si-B ordering on tetrahedral sites. Others are largely displacive but affect the coordination of the large cations (Pb, Ba, K, Ca). High-resolution electron microscopy suggests that the undulatory extinction characteristic of hyalotekite is due to a fine mosaic microstructure. This suggests that at least one of these transitions occurs in nature during cooling, and that it is first order with a large volume change. A diffuse superstructure observed by electron diffraction implies the existence of a further stage of short-range cation ordering which probably involves both (Pb,K)-Ba and (BeSi,BB)-BSi.

Prismatine: Revalidation for Boron-Rich Compositions in the Kornerupine Group

Kornerupine and prismatine were introduced independently by Lorenzen in 1884 (but published in 1886 and 1893) and by Sauer in 1886, respectively. Ussing (1889) showed that the two minerals were sufficiently close crystallographically and chemically to be regarded as one species. However, recent analyses of boron using the ion microprobe and crystal structure refinement, indicate that the boron content of one tetrahedral site in kornerupine ranges from 0 to 1. Kornerupine and prismatine, from their respective type localities of Fiskenaesset, Greenland and Waldheim, Germany, are distinct minerals, members of an isomorphic series differing in boron content. For this reason, we re-introduce Sauer's name prismatine for kornerupines with B > 0.5 atoms per formula unit (p.f.u.) of 22(O,OH,F), and restrict the name kornerupine sensu stricto to kornerupines with B < 0.5 p.f.u. Kornerupine sensu lato is an appropriate group name for kornerupine of unknown boron content. Kornerupine sensu stricto and prismatine from the type localities differ also in Fe2+/Mg ratio, Si - (Mg + Fe2+ + Mn) content, Al content, F content, colour, density, cell parameters, and paragenesis. Both minerals formed under granulite-facies conditions with sapphirine and phlogopite, but kornerupine sensu stricto is associated with anorthite and homblende or gedrite, whereas prismatine is found with oligoclase (An9-13), sillimanite, garnet, and/or tourmaline. Occurrences at other localities suggest that increasing boron content extends the stability range of prismatine relative to that of kornerupine sensu stricto.

Variations of clinopyroxene/melt element partitioning during assimilation of olivine/peridotite by low-Mg diorite magma

The effects of crystal chemistry and melt composition on the control of clinopyroxene/melt element partitioning (D) during the assimilation of olivine/peridotite by felsic magma have been investigated in Mesozoic high-Mg diorites from North China. The assimilation resulted in significant increase of Mg, Cr and Ni and only slight (< 30%) decrease of incompatible elements of the magma, and the compositional variations have been mirrored by the normally and reversely zoned clinopyroxene microphenocrysts formed at the early stage of the magma evolution. The Mg# [100 × Mg / (Mg + Fe)] values of the reversely zoned clinopyroxenes increase from 65 to 75 in the core to 85–90 in the high-Mg midsection, and reduce back to 73–79 at the rim. Trace element profiles across all these clinopyroxene domains have been measured by LA-ICP-MS. The melt trace element composition has been constrained from bulk rock analyses of the fine-grained low- and high-Mg diorites. Clinopyroxene/melt partition coefficients for rare earth elements (REE) and Y in the high-Mg group zonings (Mg# > 73–79, DDy < 1.2) are positively correlated with tetrahedral IVAl and increase by a factor of 3–4 as tetrahedral IVAl increases from 0.01 to 0.1 per formula unit (pfu). These systematic variations are interpreted to be controlled by the clinopyroxene composition. In contrast, partition coefficients for low-Mg group zonings (Mg# < 75–79, DDy > 1.2) are elevated by up to an order of magnitude (for REE and Y) or more (for Zr and Hf) at similar IVAl, indicating dominant control of melt composition/structure. DZr and DHf show a larger sensitivity to the compositional change of crystal and melt than DREE. DTi values for the low- and high-Mg zonings show a uniform dependence on IVAl. DSr and DLi are insensitive to the compositional change of clinopyroxene and melt, resulting in Sr depletions in the clinopyroxene zonings with elevated REE without crystallization of plagioclase. Our observations show that crystal chemistry and melt composition/structure may alternatively control clinopyroxene/melt partitioning during the assimilation of peridotite by felsic magma, and may be useful for deciphering clinopyroxene compositions and related crust–mantle processes.

UHP Metamorphism Documented in Ti-chondrodite- and Ti-clinohumite-bearing Serpentinized Ultramafic Rocks from Chinese Southwestern Tianshan

The southwestern Tianshan (China) metamorphic belt records high-pressure (HP) to ultrahigh-pressure (UHP) conditions corresponding to a cold oceanic subduction-zone setting. Serpentinites enclosing retrogressed eclogite and rodingite occur as lenses within metapelites in the UHP unit, which also hosts coesite-bearing eclogites. Based on the petrology and petrography of these serpentinites, five events are recognized: (1) formation of a wehrlite–harzburgite–dunite association in the mantle; (2) retrograde metamorphism and partial hydration during exhumation of the mantle rocks close to the seafloor; (3) oceanic metamorphism leading to the first serpentinization and rodingitization; (4) UHP metamorphism during subduction; (5) retrograde metamorphism during exhumation together with a second serpentinization. The peak metamorphic mineral assemblage of the serpentinized wehrlite comprises Ti-chondrodite + olivine + antigorite + chlorite + magnetite + brucite. A computed pseudosection for this serpentinized wehrlite shows that the Al content in antigorite is mostly sensititive to temperature but can also be used to constrain pressure. The average XAl = 0·204 ± 0·026 of antigorite (XAl = Al (a.p.f.u.)/8, where Al is in atoms per formula unit for a structural formula M48T34O85(OH)62, and M and T are octahedral and tetrahedral sites, respectively) included in Ti-chondrodite and average XAl = 0·203 ± 0·019 of antigorite in the matrix result in a well-constrained peak metamorphic temperature of 510–530°C. Peak pressures are less precisely constrained at 37 ± 7 kbar. The Tianshan serpentinites thus record UHP metamorphic conditions and represent the deepest subducted serpentinites discovered so far. The retrograde evolution occurs within the stability field of brucite + antigorite + olivine + chlorite and formation of Ti-clinohumite at the expense of Ti-chondrodite has been observed, suggesting isothermal decompression. The resulting P–T path is in excellent agreement with the metamorphic evolution of country rocks, indicating that the UHP unit in Tianshan was subducted and exhumed as a coherent block. To refine the metamorphic path of the ultramafic rocks, we have investigated the stability fields of Ti-chondrodite and Ti-clinohumite using piston-cylinder experiments. A total of 11 experiments were conducted at 25–55 kbar and 600–750°C in a F-free natural system. Combined with previous experiments and information from natural rocks we constructed a petrogenetic grid for the stability of Ti-chondrodite and Ti-clinohumite in F-free peridotite compositions. The formation of Ti-chondrodite in serpentinites requires a minimum pressure of about 26 kbar, whereas in Ti-rich systems it can form at considerably lower pressures. A key finding is that at UHP conditions, F-free Ti-chondrodite or Ti-clinohumite breaks down in the presence of orthopyroxene between 700 and 750°C, at temperatures that are significantly lower than those of the terminal breakdown reactions of these humite minerals. These breakdown reactions are an additional source of fluid during prograde subduction of serpentinites.

Methane concentration in sediment cores during METEOR cruise M92

We studied the concurrence of methanogenesis and sulfate reduction in surface sediments (0-25 cm below sea floor, cmbsf) at six stations (70, 145, 253, 407, 770 and 1024 m) along the Peruvian margin (12° S). This oceanographic region is characterized by high carbon export to the seafloor, creating an extensive oxygen minimum zone (OMZ) on the shelf, both factors that could favor surface methanogenesis. Sediments sampled along the depth transect traversed areas of anoxic and oxic conditions in the bottom-near water. Net methane production (batch incubations) and sulfate reduction (35S-sulfate radiotracer incubation) were determined in the upper 0-25 cmbsf of multicorer cores from all stations, while deep hydrogenotrophic methanogenesis (> 30 cmbsf, 14C-bicarbonate radiotracer incubation) was determined in two gravity cores at selected sites (78 and 407 m). Furthermore, stimulation (methanol addition) and inhibition (molybdate addition) experiments were carried out to investigate the relationship between sulfate reduction and methanogenesis. Highest rates of methanogenesis and sulfate reduction in the surface sediments, integrated over 0-25 cmbsf, were observed on the shelf (70-253 m, 0.06-0.1 and 0.5-4.7 mmol m-2 d-1, respectively), while lowest rates were discovered at the deepest site (1024 m, 0.03 and 0.2 mmol m-2 d-1, respectively). The addition of methanol resulted in significantly higher surface methanogenesis activity, suggesting that the process was mostly based on non-competitive substrates, i.e., substrates not used by sulfate reducers. In the deeper sediment horizons, where competition was probably relieved due to the decline of sulfate, the usage of competitive substrates was confirmed by the detection of hydrogenotrophic activity in the sulfate-depleted zone at the shallow shelf station (70 m). Surface methanogenesis appeared to be correlated to the availability of labile organic matter (C / N ratio) and organic carbon degradation (DIC production), both of which support the supply of methanogenic substrates. A negative correlation of methanogenesis rates with dissolved oxygen in the bottom-near water was not obvious, however, anoxic conditions within the OMZ might be advantageous for methanogenic organisms at the sediment-water interface. Our results revealed a high relevance of surface methanogenesis on the shelf, where the ratio between surface to deep (below sulfate penetration) methanogenic activity ranged between 0.13 and 105. In addition, methane concentration profiles indicate a partial release of surface methane into the water column as well as a partial consumption of methane by anaerobic methane oxidation (AOM) in the surface sediment. The present study suggests that surface methanogenesis might play a greater role in benthic methane budgeting than previously thought, especially for fueling AOM above the sulfate-methane transition zone.

Geochemistry and petrology of ODP Leg 121 basalts

Basement lavas from Sites 756, 757, and 758 on Ninetyeast Ridge are tholeiitic basalts. Lavas from Sites 756 and 757 appear to be subaerial eruptives, but the lowermost flows from Hole 758A are pillow lavas. In contrast to the compositional variation during the waning stages of Hawaiian volcanism, no alkalic lavas have been recovered from Ninetyeast Ridge and highly evolved lavas were recovered from only one of seven drill sites (DSDP Site 214). All lavas from Site 758 have relatively high MgO contents (8-10 wt%), and they are less evolved than lavas from Sites 756 and 757. Although abundances of alkali metals in these Ninetyeast Ridge basalts were significantly modified by postmagmatic alteration, abundances of other elements reflect magmatic processes. At Site 757 most of the lavas are Plagioclase cumulates, but lava compositions require two compositionally distinct, AhCb-rich parental magmas, perhaps segregated at relatively low mantle pressures. In addition, at both Sites 756 and 758 more than one compositionally distinct parental magma is required. The compositions of these Ninetyeast Ridge lavas, especially those from Site 758, require a source component with a depleted composition; specifically, the abundance ratios Th/Ta, Th/La, Ba/Nb, Ba/La, and La/Ce in these lavas are generally less than the ratios inferred for primitive mantle. Lavas from Ninetyeast Ridge and the Kerguelen Archipelago have very different chondrite-normalized REE patterns, with lower light REE/heavy REE (LREE/HREE) ratios in lavas from Ninetyeast Ridge. However, lavas from Sites 757 and 758 have Pb isotope ratios that overlap with the field defined by lavas from the Kerguelen Archipelago (Weis and Frey, this volume). Therefore, these Ninetyeast Ridge lavas contain more of a component that is relatively depleted in LREE and other highly incompatible elements, but have similar amounts of the component that controls radiogenic Pb isotopes. A model involving mixing between components related to a depleted source and an enriched plume source has been proposed for the oldest Kerguelen Archipelago basalts and Ninetyeast Ridge lavas. Although the incompatible element characteristics of the Ninetyeast Ridge lavas are intermediate between depleted MORB and Kerguelen Archipelago basalts, these data are not consistent with a simple two-component mixing process. A more complex model is required.

Analysis of biogenic magnetite nanoparticles

Cobalt doped magnetite (CoxFe3-xO4) nanoparticles have been produced through the microbial reduction of cobalt-iron oxyhydroxide by the bacterium Geobacter sulfurreducens. The materials produced, as measured by SQUID, x-ray magnetic circular dichroism, Mössbauer spectroscopy, etc., show dramatic increases in coercivity with increasing cobalt content without a major decrease in overall saturation magnetization. Structural and magnetization analyses reveal a reduction in particle size to <4 nm at the highest Co content, combined with an increase in the effective anisotropy of the magnetic nanoparticles. The potential use of these biogenic nanoparticles in aqueous suspensions for magnetic hyperthermia applications is demonstrated. Further analysis of the distribution of cations within the ferrite spinel indicates that the cobalt is predominantly incorporated in octahedral coordination, achieved by the substitution of Fe2+ site with Co2+, with up to 17 per cent Co substituted into tetrahedral sites.

Temperature and salinity reconstruction for sediment core GeoB3313-1

We reconstructed changes of temperature, salinity, and productivity within the southern Peru-Chile Current during the last 8000 years from a high-resolution sediment core recovered at 41°S using alkenones, isotope ratios of planktic foraminifera, biogenic opal, and organic carbon. Paleotemperatures and paleosalinities reached maximum values at ~5500 years ago and thereafter declined to modern values, whereas paleoproductivity continuously increased throughout the last 8000 years. We ascribe these long-term Holocene trends primarily to latitudinal shifts of the Antarctic Circumpolar Current (ACC). The concurrence with shifts in the position of the Southern Westerlies points to a common response of atmospheric and oceanographic circulation patterns off southern Chile. Millennial- to centennial-scale fluctuations of paleotemperatures and paleosalinities, on the other hand, lag displacements in the position of the Southern Westerlies but reveal a significant correlation to short-term temperature changes in Antarctica, indicating a high-latitude control of the ACC at these timescales.

Geochemistry on smectites in sediment cores CRP-1 and CRP-2 from the Ross Sea, Antarctica

TEM (transmission electron microscopy) observations and microanalyses on smectite microparticles in the sediments of the CRP-2A core were carried out to determine their origin (authigenic or detrital) and the source rocks. Smectites are dioctahedral and are Fe-rich members of the nontronite-beidellite series. They generally display both flaky and hairy shapes, but no large compositional difference between the two forms was observed. Flaky smectites are detrital while hairy smectites probably formed in situ through the reorganisation of previous flaky particles. The source rocks for smectites are probably represented by the McMurdo Volcanic Group to the south, but also by the Ferrar Dolerites and Kirkpatrick Basalts in the Transantarctic Mountains. CRP-2A smectites are Fe and Mg richer than those of the coeval or not coeval levels of the CIROS-I, DSDP 270 and 274 cores. The average compositions of smectite in CRP-1 and CRP-2A cores show a downcore trend toward more alluminiferous terms, which might reflect the increase of the chemical weathering processes on the continent.

Composition of phyllosilicates from hydrothermally altered basalts of DSDP Hole 83-504B, Costa Rica Rift, Pacific Ocean

Phyllosilicates occurring as replacements of olivine, clinopyroxene and interstitial materials and as veins or fracture-fillings in hydrothermally altered basalts from DSDP Hole 504B, Leg 83 have been studied using transmission and analytical electron microscopy. The parageneses of phyllosilicates generally change systematically with depth and with the degree of alteration, which in turn is related to permeability of basalts. Saponite and some mixed-layer chlorite/smectite are the dominant phyllosilicates at the top of the transition zone. Chlorite, corrensite, and mixed-layer chlorite/corrensite occur mainly in the lower transition zone and upper levels of the sheeted dike zone. Chlorite, talc, and mixed-layer talc/chlorite are the major phyllosilicates in the sheeted dike zone, although replacement of talc or olivine by saponite is observed. The phyllosilicates consist of parallel or subparallel discrete packets of coherent layers with packet thicknesses generally ranging from < 100 A to a few hundred A. The packets of saponite layers are much smaller or less well defined than those of chlorite, corrensite and talc, indicating poorer crystallinity of saponite. By contrast, chlorite and talc from the lower transition zone and the sheeted dike zone occur in packets up to thousands of A thick. The Si/(Si + A1) ratio of these trioctahedral phyllosilicates increases and Fe/(Fe + Mg) decreases in the order chlorite, corrensite, saponite, and talc. These relations reflect optimal solid solution consistent with minimum misfit of articulated octahedral and tetrahedral sheets. Variations in composition of hydrothermal fluids and precursor minerals, especially in Si/(Si+A1) and Fe/(Fe+Mg) ratios, are thus important factors in controlling the parageneses of phyllosilicates. The phyllosilicates are generally well crystallized discrete phases, rather than mixed-layered phases, where they have been affected by relatively high fluid/rock ratios as in high-permeability basalts, in veins, or areas adjacent to veins. Intense alteration in basalts with high permeability (indicating high fluid/rock ratios) is characterized by pervasive albitization and zeolitization. Minimal alteration in the basalts without significant albitization and zeolitization is characterized by the occurrence of saponite ± mixed-layer chlorite/smectite in the low-temperature alteration zone, and mixed-layer chlorite/corrensite or mixed-layer talc/chlorite in the high-temperature alteration zone. Textural non-equilibrium for phyllosilicates is represented by mixed layering and poorly defined packets of partially incoherent layers. The approach to textural equilibrium was controlled largely by the availability of fluid or permeability.

Geochemistry of low-temperatuer altered basalts of ODP Hole 136-843B

The mineralogy and chemistry of altered basalts and the stable isotopic compositions of secondary vein carbonates were studied in cores from Ocean Drilling Program Hole 843B, located in 95-Ma crust of the Hawaiian Arch. Millimeter- to centimeter-sized dark alteration halos around veins are 5%-15% altered to celadonite and Fe-oxyhydroxides, plus minor saponite and calcite. Adjacent gray host rocks are about 15% altered to saponite and calcite. The dark halos are enriched in H2O+, CO2, FeT, K2O, MnO, and Fe3+/FeT and depleted in SiO2, Al2O3, MgO, and TiO2 relative to gray host rocks. Brown alteration halos occur around veins where veins are more abundant, and are similar to dark halos, but contain more Fe-oxyhydroxides and exhibit greater Fe2O3T contents and higher Fe3+/FeT. Stable isotopic compositions of vein carbonates are consistent with their precipitation from seawater at temperatures of 5°-40°C. Crosscutting relationships of veins and zoned vein and vesicle fillings reveal a sequence of secondary mineral formation and alteration conditions. Celadonite and Fe-oxyhydroxides formed and dark alteration halos developed relatively early, under oxidizing conditions at low temperatures (<50°C). Saponite formed later at lower seawater/rock ratios and under more reducing conditions. Calcite and pyrite formed last in veins and vesicles from more evolved, seawaterderived fluids at temperatures of 5°-40°C. A second stage of celadonite, with compositions distinct from the early celadonite, also occurred relatively late (within the "calcite stage"), and may be related to refracturing of the crust and introduction of less-evolved seawater solutions into the rocks. Trends to higher K2O contents are attributed to alteration, but high K/Ti, Ba, and Zr contents indicate the presence of enriched or transitional MORB. CO2 contents of Pacific ODP cores exhibit a general increase with age suggesting progressive fixation of CO2 as calcite in the crust, but this could be complicated by local heterogeneities in fracturing and calcite formation in the crust.

Mineralogy and geochemistry of alteration products in DSDP Leg 81 Holes

Alteration products of basalts from the four holes drilled during Leg 81 were studied and found to be characterized by the widespread occurrence of trioctahedral clay minerals (Mg smectite to chlorite). In some cases zeolites (analcite, chabazite) are associated with the saponite. A more oxidizing stage is marked by a saponite-celadonite association, presenting the geochemical characteristics of hydrothermal processes. Later stages of alteration are represented by palagonitization and subaerial weathering at two sites. These different alteration processes of basalts from Leg 81 record the paleoenvironment during the first opening stages of the Northeast Atlantic Ocean in the Paleocene-Eocene periods.