Water samples from four springs and five wells in basalt aquifers near Spring Creek National Fish Hatchery in Skamania County, Washington, were collected and analyzed for selected inorganic ions and stable isotopes. Eight samples were analyzed for carbon-14 (14C), carbon-13 ([3C), and either chlorofluorocarbons (CFCs) or tritium. This work was done to estimate the age (residence time, or time elapsed since recharge) of water issuing from springs at the hatchery.

If CFCs are present in ground water, the presence of at least a component of modem (post- 1944) water is indicated. CFC-dating suggests that ground water several hundred feet below land surface in the Underwood Heights area noah of the hatchery, including ground water discharging from the hatchery springs, contains modem water. In contrast, CFC-dating suggests that deeper ground water such as that withdrawn from the Hatchery Well may contain little or no modem water.

Concentrations of 14C in water can yield 14C-based ground-water ages, termed 14C-model ages. Unadjusted 14C-model ages (14C-model ages unadjusted for carbon mass transfers) for water discharging from the hatchery springs are on the order of several hundred years. Ground-water samples from three wells in the Underwood Heights area yielded 14C-model ages ranging from modem to several hundred years.

Unadjusted 14C-model ages for deep ground water pumped by the Hatchery Well indicate an overall age of several thousand years. However, 14C concentrations may be affected by transfers of carbon into and out of solution. The 13C values of the downgradient ground waters ranged from - 16.4 (per mil) to -18.2 per rail, isotopically heavier than expected for ground water that obtains carbon solely from root respiration in a temperate climate and undergoes no subsequent carbon-isotope fractionation or exchange. Such 13C values in the ground water near the hatchery suggest the possibility of carbon mass transfers during the evolution of these waters. Geochemical mass-transfer modeling suggests that carbon dioxide may degas and calcite may dissolve during the evolution of the modeled ground waters, but that degassing of carbon dioxide is the dominant carbon mass transfer in these ground waters. In addition, modeling results also suggest that, although some calcite may dissolve during the evolution of the water produced by the Hatchery Well, and possibly during the evolution of the shallower ground waters, the amount of calcite dissolution is small. Model testing suggests that the quantity of 14C-dead carbon added from calcite dissolution may not be sufficient to greatly affect the 14C-model ages of these ground waters. In other words, the 14C-model ages that were adjusted for various modeled carbon mass transfers are similar to the unadjusted 14C-model ages.

A comparison of CFC data with both adjusted and unadjusted 14C data suggests that water discharging at the hatchery springs contains a mixture of modem and old water, where old water is defined as water recharged prior to 1944. The CFC data support a component of modem water, whereas the 14C data suggest a component of old water. Similar results were obtained from a comparison of CFC data with adjusted and unadjusted 14C data for water collected from Well 3. Well 3 is north of the hatchery springs, on a flow path that appears to be parallel to and similar in length to the flow path leading to the hatchery springs. Water from the Hatchery Well, however, may be devoid of modem water and appears to have an overall age on the order of thousands of years.