Dating pollen concentrated from eolian sediments provides a new way to establish a chronological framework on the Loess Plateau of China. We show that pollen deposited simultaneously with sediment in a stable environment can provide reliable ages. We suggest that the reliability of pollen dating can be evaluated by comparison with wood cellulose or charcoal ages from the same stratigraphic level. Dating pollen concentrates from the various profiles indicates paleomonsoon precipitation variability at the loess/desert transitional belt from the late Pleistocene to the early Holocene.

Rivers and shallow groundwater are deep groundwater precursors. Their dissolved inorganic carbon content (DIC) and its isotopic composition are end members in the evolution of these properties in confined situations, and are therefore essential information when applying carbon isotopes as tracers of groundwater processes and determining aquifer residence times using 14C. During studies of regional aquifer systems in New Zealand, a simple model has been developed to explain the isotopic compositions of DIC encountered in rivers and shallow groundwater. The model format incorporates a diagrammatic approach, providing a framework for tracing the subsequent evolution of DIC in both precipitation-and river-recharged aquifers under closed conditions. DIC concentration of rivers continuously adjusts toward chemical and isotopic equilibrium between direct addition of CO2 to the water (via plant respiration and decay of dead organic material) and exchange of CO2 across the river-air interface. In the shallow groundwater situation, the gaseous reservoir is soil CO2, generally at significantly higher partial pressure. In both cases, calcite dissolution or other processes may be an additional source of DIC directly added to the bicarbonate and dissolved CO2 components; while these may add or remove DIC, steady-state isotopic concentrations are considered to be determined only by the dynamic balance between directly added CO2 and gas exchange. This model allows the calculation of steady states, using selectable parameters in river or groundwater situations. These appear as straight lines in 13C or 14C vs. 1/DIC, or total 14C vs. DIC plots, into which the experimental data can be inserted for interpretation. In the case of 14C, the steady-state balance is very often complicated by the presence of an old component in the directly added DIC; the understanding achieved via the 13C patterns is helpful in recognizing this. Data from four contrasting aquifer systems in New Zealand. The success of the approach has depended crucially on DIC concentrations measured very accurately on the isotope samples, rather than separate chemical analyses.

We report a practical system to mass-produce accelerator mass spectrometry (AMS) targets with 10-100 micrograms carbon samples. Carbon dioxide is reduced quantitatively to graphite on iron fibers via manganese metal, and the Fe-C fibers are melted into a bead suitable for AMS. Pretreatment, reduction and melting processes occur in sealed quartz tubes, allowing parallel processing for otherwise time-intensive procedures. Chemical and isotopic (13C, 14C) blanks, target yields and isotopic fractionation were investigated with respect to levels of sample size, amounts of Fe and Mn, pretreatment and reduction time, and hydrogen pressure. With 7-day pretreatments, carbon blanks exhibited a lognormal mass distribution of 1.44 micrograms (central mean) with a dispersion of 0.50 micrograms (standard deviation). Reductions of 10 micrograms carbon onto targets were complete in 3-6 h with all targets, after correction for the blank, reflecting the 13C signature of the starting material. The 100 micrograms carbon samples required at least 15 h for reduction; shorter durations resulted in isotopic fractionation as a function of chemical yield. The trend in the 13C data suggested the presence of kinetic isotope effects during the reduction. The observed CO2-graphite 13C fractionation factor was 3-4% smaller than the equilibrium value in the simple Rayleigh model. The presence of hydrogen promoted methane formation in yields up to 25%. Fe-C beaded targets were made from NIST Standard Reference Materials and compared with graphitic standards. Although the 12C ion currents from the beads were one to two orders of magnitude lower than currents from the graphite, measurements of the beaded standards were reproducible and internally consistent. Measurement reproducibility was limited mainly by Poisson counting statistics and blank variability, translating to 14C uncertainties of 5-1% for 10-100 micrograms carbon samples, respectively. A bias of 5-7% (relative) was observed between the beaded and graphitic targets, possibly due to variations in sputtering fractionation dependent on sample size, chemical form and beam geometry.

We present radiocarbon determinations for 271 New Zealand archaeological samples measured at the University of Waikato Radiocarbon Dating Laboratory between 1975 and 1995. A discard protocol is applied to the series and the list culled to winnow the acceptable dates from those that may incorporate error. None of the 221 acceptable 14C determinations older than 600 BP (in the case of terrestrial samples) or 930 BP (in the case of marine and estuarine shell) extends beyond cal AD 1250. This conclusion supports the short chronology model of New Zealand prehistory presented by Anderson (1991).

The number of paleoecological records for the Sierra Nevada of California has increased substantially since the compilation of Adam (1985). We examine here the geographical and temporal distribution of records within the range in order to identify areas for which "gaps" exist in our paleoecological knowledge. Seventy-two sites with paleoecological information are identified; these sites are dated with 234 radiocarbon dates. Sites occur primarily between ca. 36 degrees N and 38 degrees 30'N latitudes, and from ca. 1000 m to over 3000 m elevation on both sides of the Sierran crest, although more sites have been analyzed on the west side of the crest than the east side. In general, packrat (Neotoma) midden series are located at the lowest elevations, meadow and marsh cores originate from mid-elevations, and lake sediments have been analyzed from the highest elevations. Significant gaps in our knowledge occur for much of the east side of the crest, for both sides of the range above modern treeline, and for time periods older than the latest Pleistocene.

Reviewed by Marian Scott, Department of Statistics, University of Glasgow

We performed radiocarbon measurements using accelerator mass spectrometry (AMS) on 6 stalagmites, 3 stalactites and 7 seepage waters from four different caves in Southwest France and Belgium in order to calculate the dead carbon proportion (dcp). All the speleothems studied are modern and annually laminated, which offers the advantage of an accurate chronology, with better than one-year resolution. Coupled with the fact that very little calcite is necessary for an AMS measurement (between 1.5 and 7 yr of calcite deposit), we obtained dead carbon values within an uncertainty limit of +/1.5%. Results show that the dead carbon proportion varies from 9.2% to 21.9% for calcite deposits and from 3.6% to 21.9% for water. In each sampling site, the dcp is homogeneous. Although the inter-site dcp varies by >11%, its average value of 15.5% +/4.4 still lies within the uncertainty range of the accepted value of 15% +/5 (dilution factor of 0.85 +/0.5). We compare the average dcp of each site with the local geology, vegetation and climate. Given similar geology and temperature, the highest dcp values are found under forest cover; dcp difference is up to 9%. However, the Belgian site, which is also under a forest, shows a dcp very close to the dcp found under grassland sites of Southwest France, which proves that other unknown factors may play an important role in dissolution processes. Secondary calcite deposition and redissolution in the soil zone or more likely in the fracture system before reaching the cave itself could also explain the inter-site differences. The IAEA isotopic model (Pearson model adapted for open systems) is in good agreement with the measured activities.

We present a proposal for a further intercomparison exercise following discussions at the 16th International Radiocarbon Conference in Groningen in 1997. This new intercomparison will build on previous exercises by making use of both reference materials already characterized and additional known-age material. For this comparison, we describe two separate but essentially related protocols that are meant to satisfy the different priorities of radiometric and AMS laboratories. The new intercomparison is planned to begin in mid-1998.