B. Kromer

Radiocarbon in Atmospheric Carbon Dioxide and Methane: Global Distribution and Trends

For many years, there has been a growing concern in the field of atmospheric chemistry about anthropogenic and natural perturbations of the major atmospheric cycles of carbon, nitrogen and sulfur, and recently, oxygen. The concern is mainly due to the implications of these trace gases on global climate. In view of the atmospheric carbon cycle, the most abundant trace gases, carbon dioxide and methane, just recently became the subject of detailed 14C investigations. These may play an important role in providing the supplementary and independent information needed to better evaluate the current observations.

THE FIRST TRANS-ARCTIC 14C SECTION: COMPARISON OF THE MEAN AGES OF THE DEEP WATERS IN THE EURASIAN AND CANADIAN BASINS OF THE ARCTIC OCEAN

We present Δ14C data collected during three cruises to the Arctic Ocean that took place in the summers of 1987 (POLARSTERN cruise ARK IV/3), 1991 (ARCTIC 91 Expedition), and 1994 (Arctic Ocean Section 94). The cruise tracks of these three expeditions cover all major basins of the Arctic Ocean (Nansen, Amundsen, Makarov and Canada basins), and can be combined to a trans-Arctic section reaching from the Barents Sea slope to the southern Canada Basin just north of Bering Strait. The section is based on 17 stations covering the entire water column (about 250 data points). The combined Δ14C data set was produced from a mixture of large volume samples measured by low-level counting and small volume samples measured by Accelerator Mass Spectrometry (AMS). We use the Δ14C section, together with previously published Δ14C data from single stations located in several basins of the Arctic Ocean, to derive mean “ages” (isolation times) of the deep waters in the Arctic Ocean. We estimate these mean “ages” to be ≈ 250 years in the bottom waters of the Eurasian Basin and ≈ 450 years in the Canadian Basin Deep Water. A remarkable feature of the Δ14C section is the homogeneity in the 14C distribution observed in the deep Canadian Basin. Within the measurement precision of about ±2‰ (LV) to about ±5‰ (AMS), we cannot detect significant horizontal or vertical Δ14C gradients below 2000 m depth between the northern boundary of the Makarov Basin and the southern margin of the Canada Basin. There is no statistically significant difference between samples measured by AMS and by low-level counting.

Mid 1980s distribution of tritium, 3He, 14C, and 39Ar in the Greenland/Norwegian seas and the Nansen Basin of the Arctic Ocean

Abstract The distributions of tritium 3 He , 14C and 39Ar observed in the period between 1985 and 1987 in the Greenland/Norwegian Seas and the Nansen Basin of the Arctic Ocean are presented. The data are used to outline aspects of the large-scale circulation and the exchange of deep water between the Greenland/Norwegian Seas and the Nansen Basin. Additionally, semi-quantitative estimates of mean ages of the main water masses found in these regions are obtained. Apparent tritium 3 He ages of the upper waters (depth 1,500m depth) of the Greenland/Norwegian Seas show apparent tritium 3 He ages between about 17 years in the Greenland Sea and 30 years in the Norwegian Sea. 39Ar based estimates of the Nansen Basin intermediate, deep and bottom water ages are 91+26−23, 161+50−44 and 277+33−31 years for Arctic Intermediate Water (AIW), Eurasian Basin Deep Water (EBDW) and Eurasian Basin Bottom Water (EBBW), respectively. Within the errors, age estimates of EBDW and EBBW based on 14 C tritium correlations are consistent with those derived from 39Ar (163 to 287 years for EBDW and 244 to 368 years for EBBW). A quantitative evaluation of the data in terms of deep water formation and exchange rates based on box model calculations is presented in an accompanying paper.

The distribution of 14C and 39Ar in the Weddell Sea

Carbon 14 and 39Ar data from the Weddell Sea are presented and discussed. Values of Δ 14C and 39Ar are low in the winter mixed layer (Δ 14C ≈ −90 to −125‰; 39Ar ≈ 85 % modern). These low values are consistent with the surface layer dynamics which is dominated by entrainment of relatively old water of circumpolar origin and reduced gas exchange during sea ice cover. The Δ 14C and 39Ar values of the deep and bottom waters range from −160 to −150‰ and 38 to 57% modern, respectively. The Δ 14C values of Weddell Sea Bottom Water (WSBW) found in the central Weddell Sea along a 0° longitude section are only slightly higher than those of the overlying Weddell Sea Deep Water (WSDW) showing that the influence of bomb 14C on these waters is small. Part of the WSBW with higher Δ 14C values observed in the northwestern Weddell Sea seems to escape through the South Sandwich Trench, and part seems to mix from a boundary current into the central Weddell Sea. The observed 14C distribution is consistent with the hypothesis that Ice Shelf Water (ISW) is a source of WSBW. A simple conceptual model of the surface layer dynamics is used to estimate the prebomb Δ 14C values of Surface Water and Winter Water to be about −140 and −130‰, respectively. Using mixing ratios between WSDW and shelf water derived from temperature/salinity and 3He data, the prebomb Δ 14C values of WSBW are estimated to be −157‰ (potential temperature of WSBW: −0.7°C). The 39Ar concentration of WSBW with a potential temperature of −0.7°C is determined to be 57% modern. Bomb radiocarbon water column inventories are estimated and discussed.

AMS 14C measurement of small volume oceanic water samples: Experimental procedure and comparison with low-level counting technique

Abstract The technique for small volume oceanic AMS 14 C measurement is described. The procedure includes sampling, CO 2 extraction from the water samples, target preparation and measurement at the ETH/SIN AMS facility. AMS 14 C data from a station in the southern Weddell Sea with an accuracy of ± 5‰ are presented. The data are in good agreement with large volume 14 C measurements done by conventional low-level counting techniques with an accuracy of ± 2‰

High-precision measurement of oceanic 226Ra

Abstract A system capable of oceanic 226 Ra measurements with a precision of ±1% is described, which represents an improvement of approximately a factor of three over existing techniques. 222 Rn grown-in from 226 Ra decay in 14-l seawater samples is quantitatively transferred to, and measured in, proportional gas counters. Errors other than counting statistics are estimated not to exceed ±0.5%, which is consistent with repeated 226 Ra measurements on the same samples. A NE Atlantic 226 Ra depth profile (2000–5000 m) is reported as an example. It is found that with the precision reported here, certain hitherto unresolved features of the 226 Ra distribution in deep water become apparent.

Electronics for low-level counting using a microcomputer

Abstract A low-cost electronic system for low-level radioactivity measurements is described. Counts of one or more detectors are handled by integrated amplifiers/quad-discriminators and anticoincidence units, interfaced to an 8-bit microcomputer with external printer which provide a maximum of 15 recording channels. Software effects a clock and counting timers, and furthermore data manipulation such as output tabulation and testing of counting statistics. Interrupt-controlled microprocessopr input is realized at 80 μs electronic dead time. This system was implemented to adopt three gas counter detectors, each equipped with its own guard counter, and with five recording channels per detector, and it has performed extremely well over more than a year of continuous operation. Its expected count rate loss of less than 0.2% has been verified by observation.