In fractured rock aquifers, apparent groundwater ages obtained with environmental tracers (e.g.,14C, CFC-12, and 3H) usually do not represent the hydraulic age of the water. Diffusion of solute between the fractures and matrix results in apparent ages that are greater than hydraulic ages, and that may be different for different tracers. We use approximate analytical solutions and numerical simulations of tracer transport through fractured porous media to illustrate the dependence of 14C and CFC-12 ages and 3H concentrations on fracture and matrix properties. In the Clare Valley, South Australia, environmental tracer data are interpreted in conjunction with hydraulic data to constrain flow parameters in a fractured shale aquifer. Hydraulic conductivity, matrix porosity, fracture spacing, and groundwater age are measured, and a value for matrix diffusion coefficient is assumed. Equations describing tracer distribution and hydraulic properties of the system are solved simultaneously, to yield estimates of fracture aperture, vertical water velocity, and aquifer recharge rate. In particular, the recharge rate is estimated to be approximately 100 mm yr-1. A sensitivity analysis showed that this value is most sensitive to the measured values of matrix porosity and groundwater age, and highly insensitive to the measured hydraulic conductivity and the assumed matrix diffusion coefficient. A major horizontal fracture at 37 m depth intercepts most of the vertical flow. The leakage rate to the deeper flow system is estimated to be less than 0.1 mm yr-1.