4.6 Discussion of Tracer‑Specific Issues

The [latex]_{}^{3}H[/latex]/[latex]_{}^{3}{He}[/latex] age‑dating method has some advantages over other transient tracers for young groundwater. For instance, the [latex]_{}^{3}H[/latex]/[latex]_{}^{3}{He}[/latex] age from Equation (14) is theoretically independent of a transient historical record of atmospheric concentrations, unlike the [latex]SF_{6}[/latex], CFC, or [latex]_{}^{3}H[/latex]‑only methods. However, the extreme variation of [latex]_{}^{3}H[/latex] in the mid‑twentieth century—the bomb peak, as shown in Figure 7—can lead to high [latex]_{}^{3}{He}_{trit}[/latex] values that are more conducive to diffusion and dispersive mixing of [latex]_{}^{3}{He}_{trit}[/latex] during transport through the aquifer (Scanlon et al., 2002; Solomon & Sudicky, 1991). These effects are much more likely in the northern hemisphere than in the southern hemisphere where the bomb peak was much lower.

Compared to transient tracers like [latex]SF_{6}[/latex], where more recent atmospheric concentrations are much higher than background concentrations—as discussed in Section 5—the [latex]_{}^{3}{He}_{trit}[/latex] signal in very young groundwater is quite small. This can be observed on the left-hand side of Figure 13 where [latex]_{}^{3}{He}_{trit}[/latex] values are much lower than [latex]_{}^{3}H[/latex]. The analytical detection limit for[latex]_{}^{3}H[/latex] is very low, but the small [latex]_{}^{3}{He}_{trit}[/latex] is modeled from noble gas data and determined by subtraction as shown in Equation (18). Modeled [latex]_{}^{3}{He}_{trit}[/latex] that is slightly negative (e.g., negative, with absolute value of < 0.5 TU) is occasionally calculated when groundwater is very young. However, large negative values could indicate an issue with sampling, analysis, or noble gas modeling. Extreme cases are shown in Figure 18 and Figure 19.

Mixing of old and young groundwater sources may also affect [latex]_{}^{3}H[/latex]/[latex]_{}^{3}{He}[/latex] tracer age. While groundwater mixtures are a complex topic that may be best evaluated with numerical models, [latex]_{}^{3}H[/latex]/[latex]_{}^{3}{He}[/latex] ages tend to be biased toward the younger groundwater source. The bias may be minor for moderate mixing. For example, a binary mixture of Source 1 (3.2 years) and Source 2 (25 years) in equal proportions would appear to have a [latex]_{}^{3}H[/latex]/[latex]_{}^{3}{He}[/latex] age of 11 years, while the average age is 14 years (Table 5). An extreme case is a binary mixture of pre‑bomb peak groundwater with very young groundwater. In this case, a mixture of 50 percent 3.2‑year‑old groundwater and 50 percent 73‑year‑old groundwater would appear to have a [latex]_{}^{3}H[/latex]/[latex]_{}^{3}{He}[/latex] age of about ten years, which is much younger than the average age of the two sources (38.2 years; Table 5).

As discussed previously, the solubilities of [latex]He[/latex] and [latex]Ne[/latex] are low and similar. If ∆[latex]Ne[/latex] is negative—which indicates degassing, as illustrated in Figure 15, Step 2b—noble gas modeling for the determination of [latex]_{}^{3}{He}_{trit}[/latex] becomes more complex. Since [latex]_{}^{3}{He}[/latex] is a decay product, it is important to know when the degassing occurred. If degassing occurs near the point of recharge, then very little [latex]_{}^{3}{He}_{trit}[/latex] would be lost—it has not been produced yet at the time of degassing—and only [latex]_{}^{3}{He}_{atm}[/latex] would be lost from the sample. Correcting [latex]_{}^{3}{He}_{trit}[/latex] for degassing (e.g., with the CE model) would lead to an older groundwater age estimate than the actual groundwater age. If degassing occurs near the point of discharge or during sampling, then [latex]_{}^{3}{He}_{trit}[/latex] would be lost from the sample. In this case, correction of [latex]_{}^{3}{He}_{trit}[/latex] for degassing is necessary to achieve the best estimate of groundwater age. However, these corrections are complicated because the original recharge temperature and excess air are also uncertain due to the degassing. One approach that has been used to reduce the number of unknowns is to assume a reasonable recharge temperature and excess air value—based on other groundwater samples collected in the area—and then apply the CE model to correct for degassing (Gilmore, Genereux, Solomon, & Solder, 2016).

Two additional conditions may cause challenges when using the [latex]_{}^{3}H[/latex]/[latex]_{}^{3}{He}[/latex] method: very deep water tables (i.e., thick vadose zone) and/or very low recharge rates. A research site with a deep water table and low recharge rate may preclude the use of [latex]_{}^{3}H[/latex]/[latex]_{}^{3}{He}[/latex] if the bulk of atmospheric [latex]_{}^{3}{He}[/latex] has decayed prior to recharge. The vadose zone transport time is calculated as shown by Equation (22).

where:

Equation (22) can be used in many cases for a rough estimate of whether vadose zone transport is likely to exceed the applicable range for nearly complete decay of atmospheric levels of [latex]_{}^{3}H[/latex]. Or, in the case of a shallower water table where recharge rate is very low, diffusion of [latex]_{}^{3}{He}_{trit}[/latex] from groundwater to the gas phase in the vadose zone may dominate over the advection of [latex]_{}^{3}{He}_{trit}[/latex] into the groundwater system. The [latex]_{}^{3}{He}_{trit}[/latex] signal can be lost if recharge rates are about 3 mm yr^-1 or lower. In contrast, [latex]_{}^{3}{He}[/latex] loss is likely less than 20 percent when the recharge rate is 30 mm yr^-1 or higher (Gilmore et al., 2021, and references in their Table 2; Solomon et al., 1993). In dual porosity settings including fractured porous rock aquifers, differential transport of ³H and ³He may occur due to differences in aqueous diffusion coefficients (LaBolle et al., 2006). Because ³H and ³He have different diffusion coefficients, they can be separated during transport. In such system, the tracer age may not have a simple relationship to the fluid velocity.