Page 150 - Hydrogeology Principles and Practice
P. 150
HYDC04 12/5/05 5:36 PM Page 133
Environmental isotope hydrogeology 133
14
ratio along a groundwater hydraulic gradient rep- emission of a β particle. In a similar way to C
36
resents Cl decay, and after correcting for different production, tritium is produced naturally mainly in
36
sources of Cl, Cl ages can be estimated. In an ideal the upper atmosphere by interaction of cosmic ray-
36
situation, if Cl and Cl are solely derived from atmo- produced neutrons with nitrogen. After oxidation
3
1
spheric sources with no internal sources or sinks, to H HO, tritium becomes part of the hydrological
36
except for Cl decay and nucleogenic production, cycle. Analysis of tritium requires distillation fol-
36
and if the initial Cl/Cl ratio (R ) and the secular lowed by electrolytic enrichment of the tritium con-
o
36
equilibrium Cl/Cl ratio (R ) can be estimated, then tent. The enriched sample is converted to ethane and
se
groundwater age estimates can be determined by the gas scintillation techniques are used to measure the
following equation (Bentley et al. 1986): tritium content. Combined field and laboratory
errors yield an accuracy of ±2 TU or better.
−
t =− 1 log R R se eq. 4.16 Natural levels of tritium in precipitation are estim-
−
λ e R R ated to be between 0.5 and 20 TU. The tritium con-
36 o se
centrations of four long-term precipitation records
36
where λ is the decay constant for Cl, and R is the are shown in Fig. 4.9. The records illustrate the effect
36
36
Cl/Cl measured in groundwater. of atmospheric testing of thermonuclear devices
Love et al. (2000) presented two modifications between 1952, prior to which there were no measure-
of equation 4.16 to allow for the addition of Cl via ments of natural tritium levels in the Earth’s atmo-
leakage or diffusion from an adjacent aquitard. In sphere, and the test ban treaty of 1963 when tritium
36
their study, Love et al. (2000) applied the Cl dating concentrations, as a result of nuclear fusion reactions,
technique to groundwaters of the south-west flow reached a peak of over 2000 TU in the northern hemi-
system of the Great Artesian Basin (see Box 2.11) in sphere. A clear seasonal variation is also evident with
north-east South Australia. The main aquifer system measured tritium concentrations less in the winter
comprises Cretaceous gravels, sands and silts of the compared with the summer when tritium is rained
Jurassic-Cretaceous aquifer, with mean flow velocit- out of the atmosphere. In the southern hemisphere
ies, calculated from the rate of decrease of absolute (and at coastal sites in general), the greater influence
36 −1
Cl concentrations, of 0.24 ± 0.03 m a . Calculated of the oceans leads to a greater dilution with water
36
Cl ages of the confined aquifers, although complic- vapour resulting in lower tritium concentrations
ated by addition of Cl via diffusion from the overlying overall. Currently, atmospheric background levels in
aquitard, range from 200 to 600 ka (Table 4.2; Love the northern hemisphere are between about 5 and
et al. 2000). Groundwater trends, from the eastern 30 TU and in the southern hemisphere between 2 and
margin of the Basin in north-east Queensland towards 10 TU (IAEA/WMO 1998).
the centre of the Basin in South Australia, show a The application of the dating equation (eq. 4.8)
36
decrease in the Cl/Cl ratio in the direction of the with tritium data to obtain groundwater ages is prob-
hydraulic gradient from approximately 100 × 10 −15 to lematic given the variation in the initial activity, A .
0
about 10 × 10 −15 over a distance of 1000 km indicat- Ambiguity arises in knowing whether the input
ing ages approaching 1.1 Ma near the end of the concentration relates to the time before or after the
flowpath (Bentley et al. 1986). 1963 bomb peak. Nevertheless, tritium concentra-
tions provide a relative dating tool with the presence
of tritium concentrations above background concen-
4.4.4 Tritium dating trations indicating the existence of modern, post-1952
14
water. Again, as with C dating, mixing between dif-
3
Tritium ( H or T), the radioisotope of hydrogen, has ferent water types with different recharge histories
a relative abundance of about 0–10 −15 %. The unit of complicates the interpretation. Some authors (for
measurement is the tritium unit (TU) defined as 1 example, Downing et al. 1977) have attempted to use
atom of tritium occurring in 10 18 atoms of H and the amount of tritium in groundwater to correct the
−1
−1
equal to 3.19 pCi L or 0.118 Bq L . The half-life of 14 C content of a sample for the effects of dilution with
3
3
tritium is 12.38 years for H decaying to He with the modern water.