The tree-ring program at Belfast originally aimed at the construction of a 6000-year oak chronology. The stimulus for this work came from the large numbers of sub-fossil oaks uncovered in Northern Ireland during land drainage and motorway construction in the late 1960's (Pilcher et al 1977). It became clear that any attempt to build such a long chronology would break naturally into two distinct units. One unit related to the construction of a prehistoric (BC era) chronology dependent on the sampling of large numbers of essentially random sub-fossil timbers. For this unit to be successful, timbers would have to survive relatively uniformly through time. The second chronology building unit was related principally to the AD era, with a natural extension into the first millennium BC at least. This unit was envisaged as the link between the present day and the necessarily floating sub-fossil chronologies. This AD chronology was based on modern, historic, and archaeologic timbers.

Bi-decade samples of dendrochronologically matched Irish Oak, measured with a precision of ca ± 20 years, covering the period 200 to 4000 BC are presented. The data are compared with the published data of Suess, de Jong, and Mook to provide a general calibration of the 14C time scale for this period. Although the dendrochronologic sequences presented are not absolutely tied to present, the best fit (based on 14C evidence) of the Belfast data to absolute chronologies, the error and evidence associated with such positioning is given. The intervals chosen for analysis were 20 years, reducing slightly the resolution of short-term variations when compared to 10-year intervals, which are sometimes measured. However, this calibration would suffice for most scientific purposes and certainly for the calendrical conversion of 14C dates derived from archaeologic samples.

The high-precision 14C measurement of bi-decade and decade samples of Irish Oak for the time period AD 50 to 1830 is presented. The samples were taken from dendrochronologically dated Irish Oak providing an absolute chronology for this period. While the natural atmospheric 14C concentration shows cyclic deviations from a constant value, the amplitudes of such deviations vary considerably. Repeated measurement and interlaboratory checks justify a claim to accuracy in the data given. Comparison of a large number of data sets between Belfast and Seattle show that the different techniques of scintillation counting of 14C6H6 and gas counting of 14CO2 derived from different wood species grown in different continents give no significant bias. Thus, it is justifiable to use this data set as a high-precision calibration curve for the AD period. Over most of this period bi-decade samples were measured giving a slightly reduced resolution of short-term variation when compared to the measurement of decade intervals; however, any difference is not apparent when such curves are compared. It is of much greater importance that a comparison of data sets is without bias if a combined calibration curve is to have sufficient integrity for general use. Averaging the data from Belfast and Seattle would improve the validity of such an AD 14C calibration curve since it would tend to smooth out slight local variations and become internationally more acceptable.

The Hohenheim Tree-Ring Laboratory has extended the Holocene oak tree-ring chronology back to prehistoric times by analyses of subfossil tree trunks from gravel deposits along the rivers of central Europe. Hundreds of subfossil oaks can be collected each year because of widespread gravel quarrying. Despite this nearly continuous source of samples (at present, 2200 trees are analyzed), even within these deposits some limitations do exist in linking together a Holocene tree-ring sequence.

Forty-five tree rings (1881–1925) were taken from a white spruce grown near Campbell River in Mackenzie Delta, Canada, for the measurement of 14C activity variation. Because of the narrowness of the rings, 2 and sometimes 3 rings were combined to yield a total of 21 specimens. The 14C content in these specimens was measured with a liquid scintillation-PM tube counter system of the History Department of Peking University. The data points exhibit a 10‰ variation, anti-correlated with sunspot numbers. The physical implication is discussed.

Parallel determinations of 14C and ionium dates on a stalagmite from the Cango Caves provide evidence of variations in the 14C content of the atmosphere beyond the range of the California tree-ring sequence. During the Holocene growth period the 230Th dates are compatible with the tree-ring calibrated 14C dates. At 18,000 BP and between 30,000 and 40,000 BP the 14C ages are markedly younger than the 230Th ages, suggesting that the 14C level of the atmosphere was considerably higher at these times. Between the 230Th ages of 35,000 and 29,000 BP the 14C ages remain nearly constant at 29,500 BP, indicating that 14C production must have been drastically reduced during this period. The 14C fluctuation is greater than that predicted by Barbetti (1980) but it may be explained by postulating a substantial increase in the geomagnetic dipole field, for which there is mounting evidence.

The AD time series of Δ14C, δ13C, and cosmic ray fluxes, Q, were analyzed for similarities. Two cases of significant correlation between Q and tree-ring δ13C were encountered, for which up to 25% of the variance can be attributed to changes in the tree's isotope fractionation that related to solar induced climatic changes. However, it is possible that the demonstrated correlation is fortuitous because actual climate proxy records generally do not correlate significantly with the Q record.

The history of the earth's magnetic field is preserved in the fossil magnetism of archaeologic specimens, natural rocks and sediments. Samples such as lava flows and baked sherds that acquired a thermoremanent magnetization on cooling can be used to estimate ancient geomagnetic field intensities and directions. Paleofield directions can also be obtained from fine-grained sediments that acquired detrital magnetic remanence when deposited. Study of the earth's magnetic field over the last few tens of thousands of years yields information on geomagnetic dynamo theories, causes of fluctuations in cosmic-ray activity, and the formulation of a new regional chronologic tool.

The geomagnetic field is one of the major physical fields of the earth. Because its source is fluid motion in the outer core, it exhibits temporal changes, called secular variation, which are quite rapid compared to most geologic phenomena. The prehistoric secular variation is usually inferred from paleomagnetic data. We will discuss here how changes in the atmospheric 14C content can be used to gain additional insight into the behavior of the dipole moment over the past 8500 years. By rewriting the differential equations representing the 14C geochemical cycle in finite-difference form, we are able to convert the atmospheric 14C activity record into an equivalent radiocarbon dipole moment.

Relatively precise quantitative observations of geophysical parameters are available to evaluate the fluctuations of atmospheric 14C activity during the past three centuries. As reviewed by Damon, Lerman, and Long (1978), these fluctuations seem to result from three factors: 1) changes in the earth's dipole magnetic field intensity, which has been decreasing since the first measurements by Gauss (McDonald and Gunst, 1968); 2) solar modulation of the cosmic-ray production, which has been correlated with the sunspot record of Waldmeier (1961), and more recently, to the Aa geomagnetic index by Stuiver and Quay (1980); and 3) the combustion of fossil fuels (Suess, 1955). A relationship between the climatic time series and the 14C-derived record of solar change has not yet been demonstrated (Stuiver, 1980).

Previous investigations on short-term 14C variations in tree rings are compared with 14C measurements in wine samples. The comparison is made for 4 solar cycles (1903–1944) with the same method of statistical evaluation of measured results. The average amplitude of Δ14C variations as observed by various authors in tree-ring samples is ca 2 ± 1 ‰; however, wine samples show an average amplitude of 4.3 ± 1.6 ‰. The anticorrelation dependence of Δ14C on Wolf sunspot numbers was observed with a time shift between W maxima and Δ14C minima of 3–5 yr for different solar cycles.

When used with 14C records, an accurate reconstruction of the 13C/12C changes of atmospheric CO2 may be a key constraint to determining the historic activity of the biosphere as CO2 source or sink. We recently detailed the derivation of an atmospheric δ13C record for 1930–1979 from Arizona juniper tree rings (Leavitt and Long, 1983). Each tree was close to a weather station, allowing us to account for the effects of climate on the tree-ring δ13C records. The resulting atmospheric δ13C trend shows an overall decrease of ca 1.3‰, but instead of decreasing exponentially over the whole 50-year period, the best-fit curve decreases exponentially to ca 1960 and then flattens out. The drop from 1956 to 1978 was ca 0.55‰, still very similar to the measured atmospheric drop over that period of 0.65 ± 0.13‰ (Keeling, Mook, and Tans, 1979).

10Be (T1/2 = 1.5·106y) is mainly produced in the atmosphere by cosmic ray spallation reactions on nitrogen and oxygen. About 70 % of the production takes place in the stratosphere. 10Be becomes attached to aerosols within a very short time. If 10Be is produced in the stratosphere some latitudinal mixing occurs. Most of the 10Be is transferred to the troposphere during spring and early summer when, mainly at median latitudes, large stratospheric air masses enter the troposphere. Tropospheric 10Be is deposited rapidly on the earth's surface by precipitation. The mean residence time of 10Be in the atmosphere is ca 1 to 2 years. 10Be removed from the atmosphere by precipitation is either preserved in snow and ice layers, in the topsoil and the biosphere, or it enters the hydrosphere (oceans and lakes), where it is transported to the sediments.

14C dating results derived from an absolutely-dated 471-year tree-ring sequence from central European oak show a trend towards somewhat older dates than those for bristleconepine tree rings of the same age, but similar to those for Egyptian historical samples. Differences visible between these trend lines are not relevant considering the standard errors proposed by Clark (1975).

Dendrochronological studies of bristlecone pine, Pinus longaeva, have produced a continuous tree-ring sequence back to 6700 BC for the White Mountains of California and to 3258 BC for east-central Nevada.

Increasing concern with nuclear waste isolation technology is leading to additional studies of naturally occurring isotopes in ground water. Such studies provide information on 1) the use of radionuclides to estimate ground-water travel times and/or residence times. This information can he an extremely useful adjunct to conventional hydrologic data in developing the understanding of regional hydrology needed in the site selection process, and 2) the use of natural radionuclides as analogues to the behavior of radionuclides of concern in nuclear waste.

Soil gas samples were obtained from the unsaturated zone at eight sites in the Great Plains. Three of these sites were sampled extensively for gas composition and carbon isotopes. Sampling equipment consisted of a nest of gas probes vertically spaced by roughly 3m at most sites, generally approaching the water table. Water wells, 10cm in diameter, were screened in the topmost layer groundwater. Inverted cattle tanks were used to collect CO2 samples from the soil surface. The major gas components were analyzed with emphasis on CO2, δ 13C, and 14C measurements. The same components were studied in groundwater samples. Higher than atmospheric CO2 concentrations were found in all soil samples. Root respiration and oxidation of organic matter were sources for the additional CO2. When lignite was present in the unsaturated zone, gaseous oxygen reacted almost completely, and CO2 levels rose to 19%. Near the surface, annual cycles in total CO2, δ 13C, and 14C were observed. 14C activities were close to present post-bomb levels at the surface and generally declined with depth. At some sites, oxidation of lignite caused decline of 14C levels to 1 or 2% of their surface value at 8m depth. Without lignite, the 14C activity remained above 50% at all depths. Concentrations of total carbon and its isotopes in ground water remained very stable throughout the study. This implies that geochemical processes in the aquifer vary on time scales longer than the seasonal effects observed in the near-surface unsaturated zone.

Data on the depth distribution of the major atmospheric gases and the abundance of gaseous 12CO2, 13CO2, and 14CO2 in the subsoil unsaturated zone have been obtained from several sites in the western Great Plains of the United States. Sample profiles range from land surface to depths of 50m. Although each site must be considered on an individual basis, several general statements can be made regarding the profiles. 1) Diffusion of these gaseous molecules through the unsaturated zone is an important transport mechanism. 2) As predicted by diffusion theory, depth profiles of the various isotopic species of CO2 differ substantially from one another, depending on individual sources and sinks such as root respiration and oxidation of organic carbon at depth. 3) In general, post-bomb (> 100% modern) 14C activities are not observed in the deep unsaturated zone, in contrast to diffusion model predictions. 4) In spite of generally decreasing 14C activities with depth, absolute partial pressures of 14CO2 in the subsoil unsaturated zone are 1–2 orders of magnitude higher than the partial pressure of 14CO2 in the atmosphere.