Understanding surface-groundwater interactions through combined physical and chemical data analysis: tracing shallow groundwater recharge in Christchurch, New Zealand

Determining sustainable water resource utilization rates is an important problem faced by regulatory agencies all around the world. One of the key parameters in determining accurate water budgeting schemes is the rate of water resource replenishment, or ‘recharge’ in groundwater systems. Fundamental questions regarding groundwater recharge include: What is the source of recharge? What is the spatial distribution of recharge? What is the annual average recharge rate, from potentially disparate sources in disparate areas? Answers to these questions can be gained through combining physical and chemical hydrogeological research tools, including stable isotopic compositions. Land-use intensification, including significant increases in dairying, has placed a priority on developing water resource management practices throughout New Zealand. In Canterbury (eastern South Island, New Zealand), applications for groundwater abstraction have increased 2-fold in the past decade. Fortunately, a wealth of long-term physical hydrogeological data for Cantberbury Plains aquifers is available through the regional government. However, basic chemical data, including potential tracers such as stable isotopic compositions, are generally absent at the regional scale. Here we present the first compilation of δ18O and δD values from individual precipitation events, local surface waters, depression springs, and groundwaters from the greater-Christchurch area. A variety of analytical methods were used in an effort to evaluate the potential use of water resource stable isotopic compositions as tracers of surface-groundwater interaction in the local hydrologic cycle. Our analysis produced several important findings: 1) Hybrid Single Particle Lagrangian Intergrated Trajectory (HYSPLIT) analysis does not yield a consistent pattern relating isotopic composition to air parcel trajectory. 2) shallow groundwater and depression spring isotopic compositions match high-altitude derived braided river waters, but not local precipitation. However, further research is needed to better constrain long-term average δ18O and δD values for precipitation. 3) long-term trends in piezometric surface elevations reflect changes in river base flow conditions; 4) short-term fluctuations, in piezometric surface elevations, correspond with significant precipitation events. In combination, these results present a compelling case for shallow groundwater recharge by surface water channel leakage in this portion of the Canterbury Plains, demonstrating the potential application of combined physical and chemical hydrologic data analysis to informing water resource management and decision-making practices.