Groundwater dilution and residence times, and constraints on chloride source, in the Mokai Geothermal System, New Zealand, from chemical, stable isotope, tritium, and 36Cl data

Abstract

The Mokai geothermal system, New Zealand, is located within a volcanic collapse structure comprised of flat‐lying silicic rocks. The deep chloride fluid rises in the south, inside the caldera margin; near the surface the fluid flows laterally more than 12 km to the north, because of the hydraulic gradient caused by 300 m of relief. The maximum measured temperature of the production wells is near 323°C at 2500 m depth, whereas temperatures in two wells on the northern margin of the upflow are <200°C, with thermal inversions at 350 and 1000 m indicating lateral flow. These two marginal wells have about 50% of the chloride of the deep fluid, indicating dilution by cold ground water. Chloride‐bearing springs discharge 6 km north of the zone of upflow and are even further diluted by cold groundwater than the wells on the margin of the system. The tritium data in this hydrologically rather simple system indicate that there is a mean residence time of greater than 100 years for the groundwater diluent; also, older water is mixed with <25% of young (thermonuclear, post‐1955) water before the mixture dilutes outflowing chloride fluid from the deep system. The deep chloride fluid has a 36Cl/Cl ratio of 2.6×10−15 and a 36Cl concentration of 106 atoms per 10−6 kg. This fluid has a lower ratio (by a factor of 90) and a higher absolute concentration (by a factor of 7) than present‐day stream water. However, the marginal wells have been diluted by groundwater with at least 500 36Cl atoms per 10−6 kg, 30 times richer in 36Cl than present‐day stream waters. This spike is most likely due to circulation of precipitation formed during thermonuclear testing to depths of 1000 m (though the component with the 36Cl spike only makes a moderate, <25%, contribution to the groundwater diluent; most of the diluent is much older, pre‐1955). The amount of 36Cl in the deep fluid at Mokai is 5–10 times less than that predicted from a calculation of the neutron fluxes in the silicic reservoir and underlying greywacke basement rocks. This indicates that the chloride was not derived from host rock leaching in the upper 3 km; rather, the fluids have interacted with more primitive rocks below the drilled depths of the Taupo Volcanic Zone, with the chloride possibly derived from a magmatic brine.