Groundwater ages represent an interaction between past climate, land surface processes and aquifer properties. A key to understanding the age of groundwater, and how it reflects past processes, is the ability to estimate groundwater ages. To assist in the determination of groundwater ages, we developed a method that interprets the concentration of multiple environmental tracers with a non-parametric model of groundwater ages. The model assumes only that similar aged water is present in the sample in similar age proportions. We applied the method to data collected in the Pilbara region of north-western Australia. Environmental tracer data consisting or Carbon-14, Argon-39, Krypton-85, Tritium, CFCs and SF6, were collected from seven mine de-watering bores between 2014 and 2015. The interpreted age distributions consisted of a number of groups that were consistent between different wells. The water present in samples consisted of a young component of water <10 years old, a component of water approximately 80 – 300 years old and water approximately 600 – 1000 years old. Whilst the exact proportions of each component were uncertain, the ages present in the sample were clearly identified. The young water component present in all the sampled bores is consistent with recent changes in the groundwater system due to pumping. The second component, water with ages between 80 and 300 years was present in five of the seven wells. Other studies in the Pilbara region have identified that this period was characterised by extreme flooding events, indicating that a large proportion of the aquifer storage was recharged under a different and more extreme climate. Our results highlight the value of groundwater age studies in understanding the extended history of groundwater resources, as hydraulic measurements only reflect changes of shorter periods.