Kenneth R. Ludwig

Calculation of uncertainties of U-Pb isotope data

Abstract Equations are derived for the estimation of errors and error correlations for various types of U-Pb isotope data, taking into account ion-beam instabilities, run-to-run variability in mass-discrimination, uncertainties in Pb and U concentrations, and uncertainties in initial-Pb and blank-Pb amount and isotopic composition. Equations are also given for the calculation of concordia intercept errors.

Continuous 500,000-Year Climate Record from Vein Calcite in Devils Hole, Nevada

Oxygen-18 (δ18O) variations in a 36-centimeter-long core (DH-11) of vein calcite from Devils Hole, Nevada, yield an uninterrupted 500,000-year paleotemperature record that closely mimics all major features in the Vostok (Antarctica) paleotemperature and marine δ18O ice-volume records. The chronology for this continental record is based on 21 replicated mass-spectrometric uranium-series dates. Between the middle and latest Pleistocene, the duration of the last four glacial cycles recorded in the calcite increased from 80,000 to 130,000 years; this variation suggests that major climate changes were aperiodic. The timing of specific climatic events indicates that orbitally controlled variations in solar insolation were not a major factor in triggering deglaciations. Interglacial climates lasted about 20,000 years. Collectively, these observations are inconsistent with the Milankovitch hypothesis for the origin of the Pleistocene glacial cycles but they are consistent with the thesis that these cycles originated from internal nonlinear feedbacks within the atmosphere-ice sheet-ocean system.

Mass-Spectrometric 230Th-234U-238U Dating of the Devils Hole Calcite Vein

The Devils Hole calcite vein contains a long-term climatic record, but requires accurate chronologic control for its interpretation. Mass-spectrometric U-series ages for samples from core DH-11 yielded 230Th ages with precisions ranging from less than 1,000 years (2σ) for samples younger than ∼140 ka (thousands of years ago) to less than 50,000 years for the oldest samples (∼566 ka). The 234U/238U ages could be determined to a precision of ∼20,000 years for all ages. Calcite accumulated continuously from 566 ka until ∼60 ka at an average rate of 0.7 millimeter per 103 years. The precise agreement between replicte analyses and the concordance of the 230Th/238U 234U/238U ages for the oldest samples indicate that the DH-11 samples were closed systems and validate the dating technique in general.

Crustal subsidence rate off Hawaii determined from 234U/238U ages of drowned coral reefs

A series of submerged coral reefs off northwestern Hawaii was formed during (largely glacial) intervals when the rate of local sea-level rise was less than the maximum upward growth rate of the reefs. Mass-spectrometric {sup 234}U/{sup 238}U ages for samples from six such reefs range from 17 to 475 ka and indicate that this part of the Hawaiian Ridge has been subsiding at a roughly uniform rate of 2.6 mm/yr for the past 475 ka. The {sup 234}U/{sup 238}U ages are in general agreement with model ages of reef drowning (based on estimates of paleo-sea-level stands derived from oxygen-isotope ratios of deep-sea sediments), but there are disagreements in detail. The high attainable precision ({plus minus}10 ka or better on samples younger than {approximately}800 ka), large applicable age range, relative robustness against open-system behavior, and ease of analysis for this technique hold great promise for future applications of dating of 50-1,000 ka coral.

Thorium-230 Ages of Corals and Duration of the Last Interglacial Sea-Level High Stand on Oahu, Hawaii

Thorium-230 ages of emergent marine deposits on Oahu, Hawaii, have a uniform distribution of ages from ∼114,000 to ∼131,000 years, indicating a duration for the last interglacial sea-level high stand of ∼17,000 years, in contrast to a duration of ∼8000 years inferred from the orbitally tuned marine oxygen isotope record. Sea level on Oahu rose to ≥1 to 2 meters higher than present by 131,000 years ago or ∼6000 years earlier than inferred from the marine record. Although the latter record suggests a shift back to glacial conditions beginning at ∼119,000 years ago, the Oahu coral ages indicate a near present sea level until ∼114,000 years ago.

Sea-level records at ~80 ka from tectonically stable platforms: Florida and Bermuda

Studies from tectonically active coasts on New Guinea and Barbados have suggested that sea level at ∼ 80 ka was significantly lower than present, whereas data from the Atlantic and Pacific coasts of North America indicate an ∼ 80 ka sea level close to that of the present. We determined ages of corals from a shallow submerged reef off the Florida Keys and an emergent marine deposit on Bermuda. Both localities are on tectonically stable platforms distant from plate boundaries. Uranium-series ages show that corals at both localities grew during the ∼80 ka sea-level highstand, and geologic data show that sea level at that time was no lower than 7–9 m below present (Florida) and may have been 1–2 m above present (Bermuda). The ice-volume discrepancy of the 80 ka sea-level estimates is greater than the volume of the Greenland or West Antarctic ice sheets. Comparison of our ages with high-latitude insolation values indicates that the sea-level stand near the present at ∼80 ka could have been orbitally forced.

Coral ages and island subsidence, Hilo drill hole

A 25.8-m-thick sedimentary section containing coral fragments occurs directly below a surface lava flow (the ∼1340 year old Panaewa lava flow) at the Hilo drill hole. Ten coral samples from this section dated by accelerator mass spectrometry (AMS) radiocarbon and five by thermal infrared multispectral scanner (TIMS) 230Th/U methods show good agreement. The calcareous unit is 9790 years old at the bottom and 1690 years old at the top and was deposited in a shallow lagoon behind an actively growing reef. This sedimentary unit is underlain by a 34-m-thick lava flow which in turn overlies a thin volcaniclastic silt with coral fragments that yield a single 14C date of 10,340 years. The age-depth relations of the dated samples can be compared with proposed eustatic sea level curves after allowance for island subsidence is taken. Island subsidence averages 2.2 mm/yr for the last 47 years based on measurements from a tide gage near the drill hole or 2.5–2.6 mm/yr for the last 500,000 years based on the ages and depths of a series of drowned coral reefs offshore from west Hawaii. The age-depth measurements of coral fragments are more consistent with eustatic sea levels as determined by coral dating at Barbados and Albrolhos Islands than those based on oxygen isotopic data from deep sea cores. The Panaewa lava flow entered a lagoon underlain by coral debris and covered the drill site with 30.9 m of lava of which 11 m was above sea level. This surface has now subsided to 4.2 m above sea level, but it demonstrates how a modern lava flow entering Hilo Bay would not only change the coastline but could extensively modify the offshore shelf.

Subsidence and volcanism of the Haleakala Ridge, Hawaii

Side-looking sonar (GLORIA) mapping has revealed a series of four arcuate bands of high sonic backscatter on the crest of the Haleakala Ridge, a major rift-zone ridge extending 135 km east of the island of Maui. Dredge recovery indicates that the shallowest of these bands is a drowned coral reef, and the deeper bands are also inferred to be coral reefs. The reefs occur above a prominent submarine bench 1500–2500 m deep on the ridge (H-terrace) that marks the shoreline at the end of vigorous shield building of Haleakala volcano when lava flows ceased crossing and reworking the shoreline. Since their growth these reefs have subsided as much as 2200 m and have tilted systematically about 20 m/km southward as a result of post-reef volcanic loading on the island of Hawaii, whose center of mass is about directly south of the Haleakala Ridge. The 234U/238U age of the dredged coral is 750 ± 13 ka, in reasonable agreement with an age of 850 ka for the underlying H terrace previously estimated from its relationship to other dated reefs to the southwest. Basalt glass fragments dredged from the Haleakala Ridge below the H terrace are tholeiitic and contain high sulfur indicative of eruption in water deeper than 200 m. Basalt glass fragments associated with the reefs above the H terrace are dominantly tholeiitic and contain intermediate sulfur contents, indicative of subaqueous eruption in shallow, near-shore conditions. One alkalic glass fragment was recovered above the H terrace. These relations indicate that the morphologic end of shield building as recorded by construction of the H terrace was not accompanyed by a change from tholeiitic to alkalic basalt; instead tholeiite eruptions continued for some time before the erupted lava became alkalic.

Berriasian (Early Cretaceous) radiometric ages from the Grindstone Creek Section, Sacramento Valley, California

The Grindstone Creek Section, Glenn County, Northern California is a sequence of hemipelagic mudstone, siltstone and sandstone interbedded with concretionary limestone and a few thin tuffs and bentonites. Two tuffs have been collected from a narrow interval of this sequence and subjected to mineralogical and isotopic analyses. UzPb isotopic analyses of zircon fractions from these volcanic horizons indicate an age of 137.1 + 1.6/−0.6 Ma. A detailed investigation has been conducted on the calcareous nannofossil stratigraphy of this section based on numerous samples with moderately preserved assemblages. The nannoflora is largely of Tethyan affinity, and allows direct correlation with the Berriasian stratotype section, with sections with published magnetostratigraphies and with a DSDP site drilled between known magnetic anomalies. The dated tuffs lie in the lower part of the upper BerriasianCretarhabdus angustiforatus Zone (Assipetra infracretacea Subzone) and within the narrow range ofRhagodiscus nebulosus. At three different sections, this subzone can be correlated with M-sequence Polarity Zones M16 and M16n. An independent magnetostratigraphic correlation is provided at DSDP Site 387, drilled between anomalies M15 and M16, where basal sediments containR. nebulosus. Buchia collected within a meter of the lower tuff lie within theB. uncitoides Zone which is Berriasian in age. The upper tuff level, which occurs 65 m above the lower tuff, is situated within the overlyingB. pacifica Zone. This zone had previously been correlated with the early Valanginian, but is clearly also partly of Berriasian age based on nannofossil stratigraphy. Our results allow an estimate of the age of the Berriasian-Valanginian and Jurassic-Cretaceous boundaries of 135.1 Ma and 141.1 Ma, respectively, and these fall within the range of, but differ significantiy from, several published time-scales.

U-Pb ages of uraniferous opals and implications for the history of beryllium, fluorine, and uranium mineralization at Spor Mountain, Utah

The U-Pb isotope systematics of uraniferous opals from Spor Mountain, Utah, were investigated to determine the suitability of such material for geochronologic purposes, and to estimate the timing of uranium and associated beryllium and fluorine mineralization. The results indicate that uraniferous opals can approximate a closed system for uranium and uranium daughters, so that dating samples as young as ∼1 m.y. should be possible. In addition, the expected lack of initial230Th and231Pa in opals permits valuable information on the initial234U/238U to be obtained on suitable samples of ≲10 m.y. age. The oldest207Pb/235U apparent age observed, 20.8 ± 1m.y., was that of the opal-fluorite core of a nodule from a beryllium deposit in the Spor Mountain Formation. This age is indistinguishable from that of fission-track and K-Ar ages from the host rhyolite, and links the mineralization to the first episode of alkali rhyolite magmatism and related hydrothermal activity at Spor Mountain. Successively younger ages of 13 m.y. and 8–9 m.y. on concentric outer zones of the same nodule indicate that opal formed either episodically or continuously for over 10 m.y. Several samples of both fracture-filling and massive-nodule opal associated with beryllium deposits gave207Pb/235U apparent ages of 13–16 m.y., which may reflect a restricted period of mineralization or perhaps an averaging of 21−and<13−m.y. periods of opal growth. Several samples of fracture-filling opal in volcanic rocks as young as 6 m.y. gave207Pb/235U ages of 3.4–4.8 m.y. These ages may reflect hot-spring activity after the last major eruption of alkali rhyolite.