High precision uranium isotope measurements may provide a means of determining the timescale associated with the transformation of bedrock to sediment, which includes the time required to mechanically break down rock into transportable fragments, the residence time of sediment particles in soils, streambeds, floodplains, dunes, and moraines, and their transport by wind, rivers and ocean currents to the site of final deposition on the seafloor or in lakes. The interpretation of variations in the 234U/238U ratios in sediments is based on a model for the disruption of the 238U decay series due to the loss of the decay product 234Th by recoil associated with the alpha decay of 238U. This paper presents the results of a study of 234U/238U ratios, as well as O, Nd and Sr isotopes, in fine-grained deep sea sediments, 0 to about 400 ky in age, cored in the North Atlantic Ocean at Ocean Drilling Program Site 984A. The sediments are largely siliciclastic, but have a significant carbonate component that varies between a few and 30 per cent by volume. The O isotope data obtained on separated foraminifera clearly show the last several glacial cycles, and thus provide a detailed temporal framework for the sediments. The Nd and Sr isotopic data show that the provenance of the sediment has oscillated, roughly but not exactly, in concert with the extent of continental ice volume, between a local source - probably volcanic rocks from Iceland - and a continental source. An unexpected finding is that the 234U/238U ratios of the siliciclastic portion of the sediment, isolated by leaching, show large and systematic variations that are correlated with glacial cycles and to some degree with sediment provenance. The U isotope variations are inferred to reflect differences in the transport time of the sediment - the time elapsed between the generation of the small sediment particles on Iceland and the continental source areas, and the time of deposition on the seafloor in the North Atlantic Ocean. The calculated transport times vary from less than 10 kyr to about 400 kyr. The long transport times for deposition during glacial times are inferred to be due to storage either as loess deposits or on continental shelves. During interglacials the far-traveled detritus is heavily diluted with sediment derived from Iceland and transported to the site of deposition within 10 kyr.

We present new isotopic and trace element data on post-shield alkalic lavas from the Hualalai, Mauna Kea, and Kohala. These small-volume eruptions, which presumably correspond to small-volume source regions in the mantle, serve as high resolution probes of geochemical heterogeneity to complement data available from shield-stage tholeiites that originate in the primary melting region. The post-shield isotopic ratios average over mantle volumes as much as 100 times smaller than those of the shield stage lavas. The post-shield volcanic vents are spread over an area of about 2500 km2 on the island of Hawaii. The locations extend about 35km on either side of the axis of the Hawaiian ridge and 60 to 110 km northwest of the centroid of the main melting anomaly (located between Kilauea and Loihi). Helium isotopic ratios were measured on olivine separates and, where present, pyroxene separates from the same samples. The Sr, Nd, and U-series isotopes were done on whole rock powders of the same samples. He isotopes range from 6-11 R/Ra, 87Sr/86Sr varies from 0.70345-0.70374, and epsilon-Nd from +5.3 to +7.4. The total range of Sr and Nd isotopic variations in these lavas is about twice that observed in the 2.84 km section of Mauna Kea drilled by HSDP, and similar to the range encompassed by Mauna Loa and Mauna Kea tholeiites excluding those erupted from ML since 30 Ka. For He the range is much smaller in the post-shield lavas than in the shield lavas. There is general SW NE asymmetry for all three isotope systems that could be viewed either as the Loa-Kea dichotomy or a reflection of the overall radial zoning of the plume. The amplitude of Sr and Nd heterogeneities is not markedly larger than in the shield sections of the volcanoes, which indicates that if there are larger amplitude variations in the plume, they are substantially smaller than the source regions of the post shield lavas. There is no evidence that the post-shield lavas are affected substantially by lithospheric interaction or that they are melted from isotopically anomalous material associated with pyroxene-rich domains.

A suite of isotopic measurements of vadose zone soil samples and groundwater has been successfully used to identify the sources of groundwater contamination at the Hanford site in south-central Washington. The isotopic composition of nitrate can be used to distinguish high-level tank waste (high d15N) and low-level process wastes (high d18O) from the relatively high background concentrations of nitrate in the groundwater at the site. The 87Sr/86Sr of strontium is typically elevated above background in areas where large volumes of water have been flushed through the vadose zone. Conversely, interaction between high-level caustic waste and feldspars in the vadose zone sediments releases low 87Sr/86Sr. Uranium isotope ratios (234U/238U, 235U/238U, 236U/238U) have been particularly useful for distinguishing different generations of nuclear fuel processing, allowing identification of waste in the vadose zone and groundwater derived from known or suspected leaks or spills. Data will be presented from a site adjacent to one of the high-level tank farms where multiple possible sources of contamination are suspected of causing high (and increasing) concentrations of 99Tc and nitrate in groundwater. The isotopic composition of nitrate from a multi-level well shows a clear example of vertical mixing of high nitrate, low-level waste with concentrated high-level waste. Sr and U isotope data are helping to define the possible source of the high-level and low-level waste components in the groundwater.