Tritium in Precipitation

Tritium (3H or T) is probably the most commonly employed radioisotope used to identify the presence of modern recharge. It is a short-lived isotope of hydrogen with a half-life of 12.43 years (Unterweger et al., 1980). It is directly incorporated into the water molecule (1H3HO or 1HTO) and so is the only radioisotope that actually dates groundwater. For most groundwaters, the strongest evidence that at least some active recharge occurs is the presence of measurable 3H.

Tritium is produced naturally by solar radiation, although much greater production accompanied the atmospheric testing of thermonuclear bombs between 1951 and 1980. Tritium’s short half-life and recent anthropogenic production makes it an excellent indicator of modern groundwater recharge. It is easily sampled. Analysis is routine, and the interpretation is not necessarily complicated.

Cosmogenic tritium

Natural tritium is formed in the upper atmosphere from the bombardment of nitrogen by the flux of neutrons in cosmic radiation, following the reaction:

The tritium thus formed combines with stratospheric oxygen to form water: Tritium decays to 3He by beta release: 3H ® 3He + b Tritium concentrations are expressed as tritium units (TU) where: and 1 TU = 0.118 Bq·kg–1 (3.19 pCi·kg–1) in water (IAEA 1983)

Natural tritium levels in precipitation are very low and represent a secular equilibrium between natural production and the combination of decay in the atmosphere plus loss to the hydrosphere and oceans. Production rates and concentrations in precipitation are a function of geomagnetic latitude, with greater production at higher latitudes. Very few measurements exist of natural, pre-bomb tritium in precipitation. Precipitation near Ottawa had about 15 TU, as measured in laboratory reagents mixed with Ottawa River water before 1951 (Brown, 1961). Kaufman and Libby (1954) used vintage wines to determined pre-bomb 3H concentrations of 3.4 to 6.6 TU for precipitation at the lower latitudes of the Naples NY, Bordeaux and Rhône regions. In Fig. 7-2 we see the concentrations of cosmogenic tritium measured in meteoric water prior to 1951, and the increases observed during the early tests of hydrogen-fusion devices.

Fig. 7-2  Tritium levels in precipitation prior to and during the earliest atmospheric tests of thermonuclear devices. Atmospheric tests: 1 — George, 75 kilotons, U.S.A.; 2 — Ivy-Mike, 10 megatons, U.S.A.; 3 — RDS-6s, 400 kilotons, U.S.S.R.; 4 — Castle series, total 47 megatons, U.S.A.; 5 — second Soviet thermonuclear test, 2 megatons.

Thermonuclear (bomb) tritium

On May 9, 1951, the world’s first thermonuclear flame was ignited by a 2H-3H-235U device code-named George. The Hagiwara-Fermi-Teller concept of hydrogen fusion had been successfully tested. On the first day of November, 1952, The Ivy-Mike shot demonstrated that megaton energy releases could be achieved by hydrogen fusion and the thermonuclear superbomb was born. The atmospheric detonation of these devices began a period of anthropogenic 3H production which raised concentrations in the stratosphere by several orders of magnitude.

Early configurations of the hydrogen bomb used the radiation of 235U-fission to compress the deuterium-tritium fuel to initiate fusion. Subsequent designs use the high neutron production of 235U-fission to split lithium deuteride, and the heat of fission to then ignite the deuterium and tritium products:

5He is not very stable (t½ = 6.7 · 10–22 s) and decays by neutron emission to stable 4He. The tremendous neutron flux at the end of this reaction chain activates atmospheric nitrogen, producing 3H. Although fission from detonation of uranium and plutonium bombs began on July 16, 1945, the neutron flux from these comparatively small and low-level fission tests, had no impact on meteoric 3H.

Atmospheric testing of nuclear devices between 1952 and 1962 generated a tremendous quantity of atmospheric tritium (Fig. 7-3). This substantial input created a tritium reservoir in the stratosphere which contaminated global precipitation systems for over four decades. A 1963 Soviet-American treaty banned the atmospheric testing of thermonuclear devices, although minor Chinese and French tests continued until 1980. The final year of megaton tests (1962) generated a huge peak, which appeared in precipitation in the spring of 1963. This 1963 peak became a marker used in many hydrological studies. Concentrations of 3H in precipitation are now largely back to natural, cosmogenic levels.

Even at the peak, anthropogenic tritium fallout was not a radiological threat to health. Canadian drinking water standards impose a limit on 3H of 7,000 Bq·kg–1, or about 60,000 TU. Minor amounts of 3H are presently released to the environment by nuclear power plants, nuclear fuel reprocessing facilities, preparation of weapons-grade nuclear material and by the manufacturers of tritiated paints and gas used in illuminating dials and emergency lighting.


Fig. 7-3 Tritium in precipitation from thermonuclear bomb tests since 1952. Tritium data for selected stations in North America and Europe, from the IAEA GNIP data base. Bomb test data summarized from various sources by Rath (1988) and Gonfiantini (1996).

The longest record of atmospheric tritium concentrations was initiated by R.M. Brown for precipitation at Ottawa, Canada, and begins in 1953 (Fig. 7-4). The IAEA has established several long-term records from its monitoring stations throughout the world. Numerous short-term records at American stations were generated during the peak input years between 1960 and 1980. All data are available at <>.

Fig. 7-4 Tritium in precipitation at Ottawa as measured in composite monthly samples (monitoring record established by R.M. Brown, AECL, Canada). Decreases from the peak in 1963 is due to attenuation in the oceans. The flattening of the decline after about 1980 and the sharp decline in 1990 likely reflect local activities in southern Ontario.

Two features should be observed in this figure; the consistent annual fluctuation in 3H, and the rapid decline from the 1963 peak. The greatest transfer of tritium from the stratosphere to the troposphere occurs during the spring in mid-latitude zones. This is due to seasonal disturbances in this boundary by displacement of the jet stream as upper level air circulation reorganizes in the spring. This "spring leak" annually recharges the hydrosphere (Fig. 7-4).

The decline of tritium in precipitation is not due only to decay, which decreases tritium levels by only 5.5% per year. A major factor is attenuation by the oceans and groundwaters, which have since become major reservoirs of thermonuclear 3H.