Niall C. Slowey

Temperature control on the incorporation of magnesium, strontium, fluorine, and cadmium into benthic foraminiferal shells from Little Bahama Bank: Prospects for thermocline paleoceanography

Surface sediments from Little Bahama Bank (LBB ), intersecting the subtropical thermocline, were used to assess the influence of temperature on the incorporation of Mg, Sr, F, and Cd into shells of benthic foraminifera. Samples were obtained from twelve ☐ cores along the southern slope of LBB, covering a temperature range of 18-4.5°C between 301 and 1585 m. We studied the composition of ten calcitic and one aragonitic species, which are often used in paleochemical reconstructions. Mg/Ca ratios decrease with increasing water depth in all benthic species, both with calcitic and aragonitic mineralogy, showing a strong correlation with water temperature. Similar decrease is seen in Sr/Ca but with no correlation with temperature. None of the benthic species studied here exhibits a depth or temperature related change in F/Ca. Similar trends are observed when using an ocean-wide dataset, which includes shallow and deep core tops (300–5000 m). We suggest that temperature is the primary control on the Mg content of benthic foraminifera. Based on inorganic precipitation experiments and thermodynamic considerations, presented here, a 30–40% decrease in the Mg distribution coefficient in calcite may be expected as a result of a temperature change from 25°C to 5°C, which is about half the observed change in LBB. A calibration curve applied to C. pachyderma data from core tops along the slope of Little Bahama Bank suggests that water temperature may be inferred from Mg/Ca ratios with an uncertainty of about ±0.8°C. Therefore, the Mg content of benthic foraminifera may provide a new, independent temperature proxy for studying shallow waters paleoceanography. The linear decrease in Sr/Ca with increasing depth is not correlated with temperature; the trend is constant from the ocean surface down to 5 km, suggesting that pressure related effects on the calcification process are a more likely explanation than post-depositional dissolution. Mg/Ca ratios in aragonitic shells of H. elegans covary with temperature, in accord with recent observations from corals. In contrast, the Sr and F chemistry of H. elegans is very different than that of corals and inorganically precipitated aragonites. The disparities between the elemental composition of biogenic and inorganic phases and the large intergeneric and interspecific differences observed both in planktonic and benthic foraminifera implicate temperature related physiological processes in regulating the coprecipitation of elements in foraminiferal shells. Our work demonstrates that Cd/Ca ratios of shallow calcitic species reflect the vertical distribution of nutrients; no significant influence of temperature on the partitioning of Cd into the shells was found. Our data extend the previous deep water calibration (Boyle, 1992), thereby allowing for the reconstruction of the nutrient chemistry of shallow thermocline waters.

Benthic foraminiferal Mg/Ca-paleothermometry: a revised core-top calibration

Core-top samples from different ocean basins have been analyzed to refine our current understanding of the sensitivity of benthic foraminiferal calcite magnesium/calcium (Mg/Ca) to bottom water temperatures (BWT). Benthic foraminifera collected from Hawaii, Little Bahama Bank, Sea of Okhotsk, Gulf of California, NE Atlantic, Ceara Rise, Sierra Leone Rise, the Ontong Java Plateau, and the Southern Ocean covering a temperature range of 0.8 to 18°C were used to revise the Cibicidoides Mg/Ca-temperature calibration. The Mg/Ca–BWT relationship of three common Cibicidoides species is described by an exponential equation: Mg/Ca = 0.867 ± 0.049 exp (0.109 ± 0.007 × BWT) (stated errors are 95% CI). The temperature sensitivity is very similar to a previously published calibration. However, the revised calibration has a significantly different preexponential constant, resulting in different predicted absolute temperatures. We attribute this difference in the preexponential constant to an analytical issue of accuracy. Some genera, notably Uvigerina, show apparently lower temperature sensitivity than others, suggesting that the use of constant offsets to account for vital effects in Mg/Ca may not be appropriate. Downcore Mg/Ca reproducibility, as determined on replicate foraminiferal samples, is typically better than 0.1 mmol mol−1 (2 S.E.). Thus, considering the errors associated with the Cibicidoides calibration and the downcore reproducibility, BWT may be estimated to within ±1°C. Application of the revised core-top Mg/Ca–BWT data to Cenozoic foraminiferal Mg/Ca suggests that seawater Mg/Ca was not more than 35% lower than today in the ice-free ocean at 50 Ma.

Evidence from U-Th dating against Northern Hemisphere forcing of the penultimate deglaciation

Milankovitch proposed that summer insolation at mid-latitudes in the Northern Hemisphere directly causes the ice-age climate cycles. This would imply that times of ice-sheet collapse should correspond to peaks in Northern Hemisphere June insolation. But the penultimate deglaciation has proved controversial because June insolation peaks 127 kyr ago whereas several records of past climate suggest that change may have occurred up to 15 kyr earlier. There is a clear signature of the penultimate deglaciation in marine oxygen-isotope records. But dating this event, which is significantly before the 14C age range, has not been possible. Here we date the penultimate deglaciation in a record from the Bahamas using a new U-Th isochron technique. After the necessary corrections for α-recoil mobility of 234U and 230Th and a small age correction for sediment mixing, the midpoint age for the penultimate deglaciation is determined to be 135 ± 2.5 kyr ago. This age is consistent with some coral-based sea-level estimates, but it is difficult to reconcile with June Northern Hemisphere insolation as the trigger for the ice-age cycles. Potential alternative driving mechanisms for the ice-age cycles that are consistent with such an early date for the penultimate deglaciation are either the variability of the tropical ocean–atmosphere system or changes in atmospheric CO2 concentration controlled by a process in the Southern Hemisphere.

Weaker Gulf Stream in the Florida Straits during the Last Glacial Maximum

As it passes through the Florida Straits, the Gulf Stream consists of two main components: the western boundary flow of the wind-driven subtropical gyre and the northward-flowing surface and intermediate waters which are part of the ‘global conveyor belt’, compensating for the deep water that is exported from the North Atlantic Ocean. The mean flow through the Straits is largely in geostrophic balance and is thus reflected in the contrast in seawater density across the Straits. Here we use oxygen-isotope ratios of benthic foraminifera which lived along the ocean margins on the boundaries of the Florida Current during the Last Glacial Maximum to determine the density structure in the water and thereby reconstruct transport through the Straits using the geostrophic method—a technique which has been used successfully for estimating present-day flow. Our data suggest that during the Last Glacial Maximum, the density contrast across the Florida Straits was reduced, with the geostrophic flow, referenced to the bottom of the channel, at only about two-thirds of the modern value. If the wind-driven western boundary flow was not lower during the Last Glacial Maximum than today, these results indicate a significantly weaker conveyor-belt component of the Gulf Stream compared to present-day values. Whereas previous studies based on tracers suggested that deep waters of North Atlantic origin were not widespread during glacial times, indicating either a relatively weak or a shallow overturning cell, our results provide evidence that the overturning cell was indeed weaker during glacial times.

A geostrophic transport estimate for the Florida Current from the oxygen isotope composition of benthic foraminifera

We present a new method for the quantitative reconstruction of upper ocean flows for during times in the past. For the warm (T>5°C) surface ocean, density can be accurately reconstructed from calcite precipitated in equilibrium with seawater, as both of these properties increase with decreasing temperature and increasing salinity. Vertical density profiles can be reconstructed from the oxygen isotopic composition of benthic foraminifera. The net volume transport between two vertical density profiles can be calculated using the geostrophic method. Using benthic foraminifera from surface sediment samples from either side of the Florida Straits (Florida Keys and Little Bahama Bank), we reconstruct two vertical density profiles and calculate a volume transport of 32 Sv using this method. This agrees well with estimates from physical oceanographic methods of 30–32 Sv for the mean annual volume transport. We explore the sensitivity of this technique to various changes in the relationship between temperature and salinity as well as salinity and the oxygen isotopic composition of seawater.

Glacial‐interglacial differences in circulation and carbon cycling within the upper western North Atlantic

We investigated glacial-interglacial changes in the circulation and carbon cycling in the western North Atlantic subtropical gyre using hydrographic data and downcore records of the stable isotopic compositions of individual shells of Bahamian benthic foraminifera. Potential temperature-salinity-depth relations show that modern thermocline (∼200–1000 m) and deep (∼1000–2000 m) waters in the Providence Channels, Bahamas, originate in the Sargasso Sea and are typical of the subtropical gyre. Gradients in the stable isotopic compositions of late Holocene Planulina and Cibicidoides species from the bank margins (∼400 to 1500 m depth) reflect temperature, nutrient, and isotopic gradients of modern subtropical gyre waters. The difference between the δ18O of glacial maximum and late Holocene foraminifera is ∼2.1‰ for the upper 900 m of the water column and ∼1.6‰ for deeper waters, indicating that these waters were ∼4°C and ∼2°C cooler, respectively, during glacial time. The glacial temperature gradient (dT/dz) was similar to today, while the base of the thermocline was ∼100 m shallower. These results differ significantly from our earlier results from multiple shell δ18O analyses, which implied upper thermocline waters were only ∼1°C cooler and dT/dz was greater during the glacial maximum. The difference occurs because bioturbation adversely affects multiple shell analyses of glacial-aged samples from shallow water depths. At all depths above 1500 m, foraminiferal δ13C are greater during the glacial maximum than the late Holocene by at least 0.1 to 0.2‰ (as much as 0.6‰ in the lower thermocline), indicating that nutrient concentrations throughout the thermocline were reduced and there was no oxygen minimum zone during the glacial maximum. This suggests greater, more uniform ventilation of the thermocline. Results of single and multiple shell δ13C analyses of glacial age foraminifera compare favorably because samples most affected by mixing correspond to water depths where the glacial-interglacial change of δ13C was small. Cooler upper ocean waters during the glacial maximum reflect cooler temperatures at the ocean surface where isopycnal surfaces outcrop, including large areas of the subtropical ocean. A shallower thermocline base is consistent with southward migration of the northern edge of the subtropical gyre or increased mode water production. Enhanced thermocline ventilation is consistent with more vigorous winds and all isopycnal surfaces outcropping in the area of Ekman downwelling.

U^Th dating of marine isotope stage 7 in Bahamas slope sediments

In order to understand the driving forces for Pleistocene climate change more fully we need to compare the timing of climate events with their possible forcing. In contrast to the last interglacial (marine isotope stage (MIS) 5) the timing of the penultimate interglacial (MIS 7) is poorly constrained. This study constrains its timing and structure by precise U^Th dating of high-resolution N 18 O records from aragonite-rich Bahamian slope sediments of ODP Leg 166 (Sites 1008 and 1009). The major glacial^interglacial cycles in N 18 O are distinct within these cores and some MIS 7 substages can be identified. These sediments are well suited for U^Th dating because they have uranium concentrations of up to 12 ppm and very low initial 230 Th contributions with most samples showing 230 Th/ 232 Th activity ratio of s 75. U and Th concentrations and isotope ratios were measured by thermal ionisation mass spectrometry and multiple collector inductively coupled plasma mass spectrometry, with the latter providing dramatically better precision. Twenty-nine of the 41 samples measured have a N 234 U value close to modern seawater suggesting that they have experienced little diagenesis. Ages from 27 of the 41 samples were deemed reliable on the basis of both their U and their Th isotope ratios. Ages generally increase with depth, although we see a repeated section of stratigraphy in one core. Extrapolation of constant sedimentation rate through each substage suggests that the peak of MIS 7e lasted from V237 to 228 ka and that 7c began at 215 ka. This timing is consistent with existing low precision radiometric dates from speleothem deposits. The beginning of both these substages appears to be slightly later than in orbitally tuned timescales. The end of MIS 7 is complex, but also appears to be somewhat later than is suggested by orbitally tuned timescales, although this event is not particularly well defined in these cores. fl 2002 Elsevier Science B.V. All rights reserved.

U-Th dating of carbonate platform and slope sediments

Absolute chronology of marine sediment beyond the 14C age range provides a test for models of climate change and has many other applications. U-Th techniques have been used for such chronology by dating corals, but extending these techniques to marine sediment is complicated by the presence of significant initial 230Th—both in detrital material and scavenged from seawater. In this study, we investigate four methods of solving the initial 230Th problem for a particular type of marine sediment—the aragonite-rich sediments of carbonate platforms and slopes. Bulk sediment U-Th analyses can be corrected for initial Th to yield ages with ≈2 to 3 kyr precision for highstand periods when sediment aragonite contents are particularly high. Uncertainty on the corrections causes inadequate precision for sediment from other periods, however. Removal of scavenged Th before analysis would enable a dramatic increase in this precision but has not proved successful despite a range of chemical leach approaches. Using heavy liquids to separate the various carbonate minerals found in Bahamas sediment enables an isochron approach to correct for initial Th, but the presence of initial Th from two sources requires correction or removal of one source of initial Th before the other is deconvolved by the isochron. Quantitative removal of detrital material before isochron analysis proves a successful approach. Such isochron data demonstrate that, although sediment remains closed to U-Th on a centimetre scale, nuclides are moved from grain to grain by α-recoil. Such intergrain exchange is expected to be observed in all sediments containing mineral grains with different U concentrations. Measured 234U/238U allows the recoil movement to be corrected and results in isochron ages with precision sometimes as low as 3 kyr. The accuracy of this approach has been proved by dating samples within the 14C age range. Sediments spanning the penultimate deglaciation have also been dated. After a small correction for bioturbation, the age for this event is found to be 135.2 ± 3.5 ka. This date is ≈8 kyr before the peak in northern hemisphere insolation and suggests that deglaciation is initiated by a mechanism in the southern hemisphere or tropics. This isochron approach shows considerable promise for dating of sediments older than this event, which will provide further information about the timing and mechanisms of global climate change.

VARIATION IN BIOTURBATION WITH WATER DEPTH ON MARINE SLOPES : A STUDY ON THE LITTLE BAHAMAS BANK

Abstract Reconstructing the paleoceanography of intermediate-depth waters is dependent on sedimentary records preserved on marine slopes. But knowledge of bioturbation processes in such slope environments is much poorer than for the deep sea. In this study, 210 Pb profiles were measured for the upper ≈20 cm of sediment from five box cores taken on the slopes of the Bahamas in water depths ranging from 400 to 1500 m. These were compared with data from the Bahamas bank tops and from elsewhere to assess the rate and depth of mixing in the slope environment. Excess 210 Pb inventories are too high to explain by water-column decay of 226 Ra and are derived largely from atmospheric fallout. Down-core 210 Pb xs profiles show an upper zone where 210 Pb xs activities decrease exponentially with depth suggesting uniform mixing, and a lower zone with patchy 210 Pb values reflecting isolated burrows. In contrast to bank-top sediment which shows 210 Pb xs mixing to depths of ≈40 cm and mixing rates of >1000 cm2/kyr, mixing of 210 Pb on the slopes penetrates to only ≈8 cm with mixing rates of 210 Pb -derived mixing activity in ocean margin sediments: Diffusive versus nonlocal mixing. J. Mar. Res. 54, 1207–1227]. A simple model is used to assess the effects of mixing on typical Bahamian slope sediments which have deposition rates and compositions which vary dramatically in response to sea-level change. Proxy records such as δ 18 O are expected to be displaced by up to 0.8 kyr in bulk sediment. And individual sediment constituents may be moved by up to 2 kyr relative to one another. These mixing effects are reasonably small and may be safely ignored for some studies. But for high-resolution records, or studies where precise age control is important, bioturbation of slope sediments should be considered.

FLUID FLOW THROUGH CARBONATE PLATFORMS : CONSTRAINTS FROM 234U/238U AND CL- IN BAHAMAS PORE-WATERS

The geometry, timing, and rate of fluid-flow through carbonate margins and platforms is not well constrained. In this study, we use U concentrations and isotope ratios measured on small volumes of pore-water from Bahamas slope sediment, coupled with existing chlorinity data, to place constraints on the fluid-flow in this region and, by implication, other carbonate platforms. These data also allow an assessment of the behaviour of U isotopes in an unusually well constrained water‐rock system. We report pore-water U concentrations which are controlled by dissolution of high-U organic material at shallow depths in the sediment and by reduction of U to its insoluble 4C state at greater depths. The dominant process influencing pore-water ( 234 U= 238 U) is alpha recoil. In Holocene sediments, the increase of pore-water ( 234 U= 238 U) due to recoil provides an estimate of the horizontal flow rate of 11 cm=year, but with considerable uncertainty. At depths in the sediment where conditions are reducing, features in the U concentration and ( 234 U= 238 U) profiles are offset from one another which constrains the effective diffusivity for U in these sediments to be1‐2 10 8 cm 2 s 1 . At depths between the Holocene and these reducing sediments, pore-water ( 234 U= 238 U) values are unusually low due to a recent increase in the dissolution rate of grain surfaces. This suggests a strengthening of fluid flow, probably due to the flooding of the banks at the last deglaciation and the re-initiation of thermally-driven venting of fluid on the bank top and accompanying recharge on the slopes. Interpretation of existing chlorinity data, in the light of this change in flow rate, constrain the recent horizontal flow rate to be 10.6 (3.4) cm=year. Estimates of flow rate from ( 234 U= 238 U) and Cl are therefore in agreement and suggest flow rates close to those predicted by thermally-driven models of fluid flow. This agreement supports the idea that flow within the Bahamas Banks is mostly thermally driven and suggests that flow rates on the order of 10 cm =year are typical for carbonate platforms where such flow occurs.