[A-HW23] Residence time of groundwater / surface water and water / mass cycle processes in watershed

[AHW23-01] Residence times of water and chemical flows in a karst spring

★Invited Papers

*Michael Kilgour Stewart¹, Uwe Morgenstern², Maksym Gusyev³, Joseph Thomas⁴ (1.Aquifer Dynamics & GNS, PO Box 30368, Lower Hutt, New Zealand, 2.GNS Science, Lower Hutt, New Zealand, 3.International Centre for Water Hazard and Risk Management (ICHARM), PWRI, Tsukuba, Japan, 4.Tasman District Council, Richmond, New Zealand)

Keywords:Mean residence time (MRT), Spring, Tritium, CFCs, Chloride

Residence times have been estimated using tritium, CFCs and stable isotopes in a large karst spring (Te Waikoropupu Springs, Golden Bay, New Zealand, Stewart and Thomas, 2008).This spring system, with its discharge of 14 m³/s, is representative of the flow paths over a large catchment. Combined with flow and chemical measurements, these lead to a steady-state (or average) model of the flows in the watershed. The model shows that the spring is fed by two different flow systems, each drawn in different amounts from three sources (high and low altitude rainfall, and river seepage). δ¹⁸O and chloride measurements identify the proportions of each of these flow systems. Monte Carlo estimation methods were then applied to determine the residence times of the spring and its two component flow systems and their uncertainties. Fig. 1 shows simulations to the tritium concentrations measured in the spring using exponential piston flow mixing models for each flow system. The mean residence time of the spring was 9.6 ± 5.0 years, and that for the component flow systems were 1.3 ± 0.7 years and 12.3 ± 6.7 years respectively.