Christina D. Gallup

The half-lives of uranium-234 and thorium-230

Abstract We have re-determined the 234 U and 230 Th half-lives to be 245,250±490 years (2σ) and 75,690±230 years (2σ), respectively. Using high precision thermal ionization mass spectrometric (TIMS) methods, we measured 234 U / 238 U and 230 Th / 238 U atomic ratios in 4 different materials that were likely to have behaved as closed systems for 106 years or more: zircons with concordant 238 U – 206 Pb , 235 U – 207 Pb , and 232 Th – 208 Pb ages, Iceland Spar, Table Mountain Latite, and aliquots of a solution of Harwell uraninite (HU-1). We calibrated the TIMS multipliers using U-500, U and Th gravimetric standards, and U double spike. Consistent 234 U / 238 U values for all measured materials and consistent 230 Th / 238 U values for all materials with the exception of our HU-1 solution support the secular equilibrium status. The new half-lives agree within error with previously determined values; however, errors in our values are generally smaller than those in the earlier determinations. Our 234 U half-life is about 3‰ higher than that commonly used in 230 Th dating laboratories and our 230 Th half-life is about 4‰ higher. 230 Th ages calculated with the new half-lives are generally older than those calculated with the previously used half-lives. The difference in age, though, is small throughout the 230 Th age range because our revised 234 U and 230 Th half-lives are offset from earlier values in the same sense (both to higher values). In the case of dating materials older than 350 ka in laboratories that rely solely on gravimetric standardization procedures, use of our decay constants and their associated errors will considerably reduce the errors in age arising from uncertainty in the decay constants.

Rapid sea-level fall and deep-ocean temperature change since the last interglacial period

We have dated Huon Peninsula, Papua New Guinea and Barbados corals that formed at times since the Last Interglacial Period, applying both 230 Th and 231 Pa dating techniques as a test of age accuracy. We show that Marine Isotope Stage (MIS) 5e ended prior to 113.1 8 0.7 kyr, when sea level was 319 m. During MIS 5b sea level was 357 m at 92.6 8 0.5 kyr, having dropped about 40 m in approximately 10 kyr during the MIS 5c^5b transition. Sea level then rose more than 40 m during the MIS 5b^5a transition, also in about 10 kyr. MIS 5a lasted until at least 76.2 8 0.4 kyr, at a level of 324 m at that time. Combined with earlier data that places MIS 4 sea level at 381 m at 70.8 kyr, our late MIS 5a data indicate that sea level fell almost 60 m in less than 6 kyr (10.6 m/kyr) during the MIS 5^4 transition. The magnitude of the drop is half that of the glacial^interglacial amplitude and approximatelyequivalent to the volume of the present-dayAntarctic Ice Sheet. During this interval the minimum average rate of net continental ice accumulation was 18 cm/yr, likely facilitated by efficient moisture transport from lower latitudes. At three specific times (60.6 8 0.3, 50.8 8 0.3, and 36.8+0.2 kyr) during MIS 3, sea level was between 385 and 374 m. Sea level then dropped to 3107 m at 23.7 8 0.1 kyr early in MIS 2, before dropping further to Last Glacial Maximum (LGM) values and then rising to present values during the last deglaciation. Times of rapid sea-level drop correspond to times of high winter insolation at low northern latitudes and high winter latitudinal gradients in northern hemisphere insolation, supporting the idea that these factors mayhave resulted in high water-vapor pressure in moisture sources and efficient moisture transport to high-latitude glaciers, therebycontributing to glacial buildup. We combined our sea-level results with deep-sea N 18 O records as a means of estimating the temperature and ice-volume components in the marine N 18 O record. This analysis confirms large deep-ocean temperature shifts following MIS 5e and during Termination I. Deep-ocean temperatures changed bymuch smaller amounts between MIS 5c and 2. Maximum temperature shift in the deep Pacific is about 2‡, whereas the shift at a site in the Atlantic is 4‡. Under glacial conditions temperatures at both sites are near the freezing point. The shift in the Atlantic is likelycaused bya combination of changing proportions of northern and southern source waters as well as changing temperature at the sites where these deep waters form.

The timing of high sea levels over the past 200,000 years.

The 230Th ages and 234U/238U ratios were determined for Barbados corals that grew during periods of high sea level within the last 200,000 years. The similarity of the initial 234U/238U ratios of some of the corals to the modern marine value suggests that these samples are pristine and that the marine 234U/238U ratio 83,000 and 200,000 years ago was within 2 per mil of the modern value. The accuracies of the 230Th ages are evaluated on the basis of the 234U/238U values and a model of the behavior of uranium and thorium isotopes during diagenesis. For the last three interglacial and two intervening interstadial periods, sea level peaked at or after peaks in summer insolation in the Northern Hemisphere. This overall pattern supports the idea that glacial-interglacial cycles are caused by changes in Earths orbital geometry. The sea-level drop at the end of the penultimate interglacial, the last interglacial, and a subsequent interstadial period lagged behind the decrease in insolation by 5,000 to 10,000 years.

Annual cycles of U Ca in coral skeletons and U Ca thermometry

Abstract We have discovered annual variations in the 238 U 40 Ca ratio of Porites coral skeletons. Measurements were made using thermal ionization mass spectrometric techniques, yielding precisions of ±2ℵ. (2σ) for 238 U 40 Ca and ±l‰ (2σ) for 88 Sr 40 Ca . Coralline aragonite subsamples weighed ∼2 mg, enabling submillimeter sampling resolution corresponding to ∼monthly temporal resolution. The annual nature of the cycles was confirmed by comparison with annual banding observable in X-radiographs. For two modern and one fossil sample, the amplitude of the U Ca variation ranges from 6 to 23%, well outside of analytical error. As annual U Ca cycles appear to be a general feature of primary coralline aragonite, the preservation of such features will be important in identifying unaltered coral for U-series dating studies. U Ca variations mimic and are in phase with annual variations in 88 Sr 40 Ca . For a given fractional shift in Sr Ca , the fractional shift in U Ca is about 6 times larger. For the two modern corals, 238 U 40 Ca is strongly anticorrelated with measured temperature, suggesting that 238 U 40 Ca has potential as a paleothermometer. If temperature is the only significant control on coralline 238 U 40 Ca , we reach the following conclusions from analyses of fossil samples: a Vanuatu sample, which grew halfway through the last deglaciation, gives U Ca temperatures 4 to 5°C below modern values. U Ca thermometry applied to published data for Barbados corals indicates that (1) temperature generally correlates with sea level, (2) glacial temperatures (stages 2 and 4) were 4 to 6°C lower than interglacial temperatures (stages 7a, 5e, and 1), and (3) temperatures rose from glacial to interglacial values early in the last deglaciation. Thermometry applied to Papua New Guinea corals indicates that (1) temperatures were 5 to 6°C lower than interglacial temperatures from ∼13 to ∼10 Ky bp , then rose to present values ∼9 Ky bp , (2) the temperature depression between ∼13 and ∼10 Ky bp is consistent with low temperatures observed in Vanuatu during the same time interval, and (3) the Papua New Guinea deglacial temperature history differs from that of Barbados. The results generally support estimates of tropical temperatures obtained from Sr Ca thermometry and snow line elevation data, but disagree with those based on foram transfer functions. A thermodynamic model suggests that coralline 238 U 40 Ca may also be sensitive to marine carbonate ion concentration, raising the possibility that some of the observed glacial-interglacial 238 U 40 Ca variation may result from glacial-interglacial carbonate ion changes. However, the key experiments that might establish a coralline 238 U 40 Ca-carbonate ion relationship have yet to be performed.

A penultimate glacial monsoon record from Hulu Cave and two-phase glacial terminations

Oxygen isotope records of three stalagmites from Hulu Cave, China, extend the previous high-resolution absolute-dated Hulu Asian Monsoon record from the last to the penultimate glacial and deglacial periods. The penultimate glacial monsoon broadly follows orbitally induced insolation variations and is punctuated by at least 16 millennial-scale events. We confirm a Weak Monsoon Interval between 135.5 ± 1.0 and 129.0 ± 1.0 ka, prior to the abrupt increase in monsoon intensity at Asian Monsoon Termination II. Based on correlations with both marine ice-rafted debris and atmospheric CH 4 records, we demonstrate that most of marine Termination II, the full rise in Antarctic temperature and atmospheric CO 2 , and much of the rise in CH 4 occurred within the Weak Monsoon Interval, when the high northern latitudes were probably cold. From these relationships and similar relationships observed for Termination I, we identify a two-phase glacial termination process that was probably driven by orbital forcing in both hemispheres, affecting the atmospheric hydrological cycle, and combined with ice sheet dynamics.

Uranium-series Dating of Marine and Lacustrine Carbonates

Of the possible uranium-series dating schemes, the most important and most widely applied to marine carbonates is 230Th dating, with 231Pa dating playing an increasingly important role. For this reason, this review will focus on these two methods. 230Th dating, also referred to as U/Th dating or 238U-234U-230Th dating, involves calculating ages from radioactive decay and ingrowth relationships among 238U, 234U, and 230Th. 232Th is also typically measured as a long-lived, essentially stable index isotope (over the time scales relevant to 230Th dating). At present 230Th dating can, in principle, be used to date materials as young as 3 years and in excess of 600,000 years (Edwards et al. 1987a, 1993; Edwards 1988; see Stirling et al. 2001 for an example of dating corals in excess of 600,000 years old). 231Pa dating, also referred to as U/Pa dating, involves calculating ages from the ingrowth of 231Pa from its grandparent 235U. At present 231Pa dating can be used to date materials as young as 10 years and as old 250,000 years (Edwards et al. 1997). 230Th dating covers all of the 231Pa time range and more, with somewhat higher precision, and is therefore the method of choice if a single method is applied. However, the combination of 231Pa and 230Th dating is of great importance in assessing possible diagenetic mobilization of the pertinent nuclides, and thereby, the accuracy of the ages (Allegre 1964; Ku 1968). Even if the primary age exceeds the 250,000 year limit of 231Pa dating, the combined methods can be used to assess the degree to which the samples have remained closed over the past 250,000 years (e.g., Edwards et al. 1997). Thus …

Drowning of the -150 m reef off Hawaii: A Casualty of global meltwater pulse 1A?

We present evidence that the drowning of the 2150 m coral reef around Hawaii was caused by rapid sea-level rise associated with meltwater pulse 1A (MWP-1A) during the last deglaciation. New U/Th and 14 C accelerator mass spectrometry dates, combined with reinterpretation of existing radiometric dates, constrain the age of the coral reef to 15.2-14.7 ka (U/Th age), indicating that reef growth persisted for 4.3 k.y. following the end of the Last Glacial Maximum at 19 ka. The drowning age of the reef is roughly synchronous with the onset of MWP-1A between 14.7 and 14.2 ka. Dates from coralline algal material range from 14 to 10 cal ka (calibrated radiocarbon age), 1-4 k.y. younger than the coral ages. A paleoenvironmental reconstruction incorporating all available radiometric dates, high-resolution bathymetry, dive observations, and coralgal paleobathymetry data indicates a dramatic rise in sea level around Hawaii ca. 14.7 ka. Paleowater depths over the reef crest increased rapidly above a critical depth (30-40 m), drowning the shallow reef-building Porites corals and causing a shift to deep- water coralline algal growth, preserved as a crust on the drowned reef crest.

Coralgal composition of drowned carbonate platforms in the Huon Gulf, Papua New Guinea; Implications for lowstand reef development and drowning

Abstract Collision between the South Bismarck plate and the northern edge of the Australian plate has produced an actively subsiding foreland basin in the western Huon Gulf. A series of drowned carbonate platforms and pinnacles are preserved on this margin due to a combination of this rapid subsidence and eustatic sea-level changes over the last 450 ka. We analyzed sedimentary and coralgal data from the platforms to better understand lowstand reef development and drowning in the Huon Gulf. The recovered limestones are divided into five main sedimentary facies: coral reef, coralline–foraminiferal nodule, coralline–foraminiferal crust, Halimeda , and planktonic foraminiferal limestones. Based on a comparison with modern analogues in the Indo-Pacific and elsewhere, we identified coral reef, deep fore-reef slope, deeper fore-reef slope, and pelagic/hemipelagic paleoenvironmental settings. An analysis of facies relationships and their paleoenvironmental meanings revealed lowstand corals reefs preserved at the top of the platforms that grew within ∼10 m of sea level. Two different coral assemblages were identified within this facies: (1) a shallow, high energy reef community characteristic of windward margins and limited to the deep platforms (1947, 2121, 2393 m), and (2) another shallow community but indicative of more moderate lower energy reef conditions and limited to the middle (1113, 1217, 1612 m) and shallow platforms (823 m). The change from high to lower energy reef growth conditions suggests that oceanographic/climatic conditions in the Huon Gulf have changed substantially through time, primarily through the closure of the Gulf as a result of tectonic rotation and uplift of the Huon Peninsula over the last 450 ka. Despite major environmental perturbations (i.e. relative sea-level and temperature changes) the platforms and the shallow water coral reefs exposed at the top have been able to re-establish themselves time and time again over the last 450 ka. We also identified two different incipient drowning scenarios influenced by the rate of relative sea-level rise. More rapid drowning in the middle and deep platforms produced a thin veneer of coralline–foraminiferal nodule and Halimeda limestones over the shallow coral reef material while the slower drowning experienced by the shallowest platforms allowed thick coralline–foraminiferal crust limestones to develop. We recognize three main stages of platform development: (1) initiation and growth characterized by shallow coral reef growth as the platforms grew close to sea level during the lowstands, (2) incipient drowning marked by a shift to coralline–foraminiferal nodule, crust and Halimeda limestones as the platforms began to drown during rapid eustatic sea-level rise and continued subsidence, and (3) the complete drowning of the platforms characterized by platform ‘turn off’, increased bioerosion, Fe–Mn precipitation and pelagic/hemipelagic sedimentation as the platform surfaces finally drop below the photic zone.

Foredeep tectonics and carbonate platform dynamics in the Huon Gulf, Papua New Guinea

HAWAII MR1 side-scan sonar and six-channel seismic reflection data reveal a history of carbonate platform growth, drowning, and back stepping in the Huon Gulf, Papua New Guinea. This is one of the few modern sites where active carbonate platform development and foredeep subsidence are linked. 230 Th methods date aragonitic shallow-water corals, recovered from a modern depth of 2000 m, at 348 ± 10 ka. This documents rapid subsidence of the Huon Gulf in response to the encroaching Finisterre Mountains at an average rate of 5.7 mm/yr for the past 348 k.y., the highest subsidence rate reported from any foredeep setting. Carbonate deposition has moved toward the foreland at an average rate of 110 mm/yr over the same period. Comparisons of the measured age with sea-level history (derived from the oxygen isotope record) suggest that the reefs may have formed during sea-level lowstands and drowned during rapid rates of sea-level rise.

Annual cycles of UCa in coral skeletons and UCa thermometry

Abstract We have discovered annual variations in the 238 U 40 Ca ratio of Porites coral skeletons. Measurements were made using thermal ionization mass spectrometric techniques, yielding precisions of ±2ℵ. (2σ) for 238 U 40 Ca and ±l‰ (2σ) for 88 Sr 40 Ca . Coralline aragonite subsamples weighed ∼2 mg, enabling submillimeter sampling resolution corresponding to ∼monthly temporal resolution. The annual nature of the cycles was confirmed by comparison with annual banding observable in X-radiographs. For two modern and one fossil sample, the amplitude of the U Ca variation ranges from 6 to 23%, well outside of analytical error. As annual U Ca cycles appear to be a general feature of primary coralline aragonite, the preservation of such features will be important in identifying unaltered coral for U-series dating studies. U Ca variations mimic and are in phase with annual variations in 88 Sr 40 Ca . For a given fractional shift in Sr Ca , the fractional shift in U Ca is about 6 times larger. For the two modern corals, 238 U 40 Ca is strongly anticorrelated with measured temperature, suggesting that 238 U 40 Ca has potential as a paleothermometer. If temperature is the only significant control on coralline 238 U 40 Ca , we reach the following conclusions from analyses of fossil samples: a Vanuatu sample, which grew halfway through the last deglaciation, gives U Ca temperatures 4 to 5°C below modern values. U Ca thermometry applied to published data for Barbados corals indicates that (1) temperature generally correlates with sea level, (2) glacial temperatures (stages 2 and 4) were 4 to 6°C lower than interglacial temperatures (stages 7a, 5e, and 1), and (3) temperatures rose from glacial to interglacial values early in the last deglaciation. Thermometry applied to Papua New Guinea corals indicates that (1) temperatures were 5 to 6°C lower than interglacial temperatures from ∼13 to ∼10 Ky bp , then rose to present values ∼9 Ky bp , (2) the temperature depression between ∼13 and ∼10 Ky bp is consistent with low temperatures observed in Vanuatu during the same time interval, and (3) the Papua New Guinea deglacial temperature history differs from that of Barbados. The results generally support estimates of tropical temperatures obtained from Sr Ca thermometry and snow line elevation data, but disagree with those based on foram transfer functions. A thermodynamic model suggests that coralline 238 U 40 Ca may also be sensitive to marine carbonate ion concentration, raising the possibility that some of the observed glacial-interglacial 238 U 40 Ca variation may result from glacial-interglacial carbonate ion changes. However, the key experiments that might establish a coralline 238 U 40 Ca-carbonate ion relationship have yet to be performed.