Dating of groundwater with tritium and 14C

Shallow groundwater can be dated with some accuracy on the basis of its bomb tritium content if the unsaturated soil cover and the aquifer itself is sufficiently homogeneous. A few examples from the Rhine valley are presented. The decrease in tritium level from the water table to a few metres below is nearly two orders of magnitude. Agreement between the measured or estimated variation of bomb tritium in rain during the past decade and the tritium found in shallow groundwater can be obtained if one takes into account that (a) practically no summer rain reaches the water table, and (b) water is mixed by diffusion. Both effects can also be observed in the soil moisture of the unsaturated soil above the water table. Carbon-14 increase in groundwater due to bombs is delayed compared to tritium, the reasons being delay in the biological system and exchange with the carbonate in the soil. Nevertheless lysimeters show a marked increase of 14C, which depends on the plant cover, being high in a plant-covered lysimeter and low in a bare one. A simple model is presented, which allows the evaluation of the influence of exchange on the 14C age obtained. It turns out that the deviation from the true age depends on the ratio of the carbonate content in the aquifer material to the carbonate content of the water, on the specific contact surface or the grain size but not on the groundwater velocity. On the basis of this model the experimental finding that 14C ages are usually in agreement with other age estimates despite the loss by exchange is plausible owing to the fact that only material of sufficiently coarse grain size can make up a reasonable aquifer. Assuming only exchange with a monomolecular surface layer of the carbonate grains one finds that the 14C age is likely to differ by not more than a factor of two in the most unfavourable case. Under natural conditions (steady state of cosmic-ray-produced 14C) the 14C content of shallow groundwater is hardly influenced at all by exchange, even within very finely grained soil. In the case where the 14C level is rising because of bomb testing, a new equilibrium with the monolayer must be established, by which the 14C increase in groundwater is delayed. The extent of carbonate exchange has been studied under laboratory conditions in a column. The delay of the 14C pulse in spiked water under the conditions chosen is such that the spike travels with approximately half the velocity of the water. From this and from the dispersion observed one can conclude that equilibrium with the surface monolayer is reached within a few hours. Further studies at elevated temperatures indicate the extent to which penetration into deep layers must be taken into account. (author)