Determining aquifer hydraulic parameters using Surface Nuclear Magnetic Resonance (SNMR) data: the Paleozoic Bonaparte Basin, East Kimberley, Australia

Nuclear magnetic resonance (NMR) is an analytical technique used in geophysical investigations where a magnetic field is induced to produce an NMR response from protons in water molecules in the subsurface. Surface nuclear magnetic resonance (SNMR) data are inverted to produce a layered earth model of water content in the uppermost~100 m of the earth’s crust. Given the quick and non-invasive nature of data-acquisition and the fact that water content data can be simply transformed to hydraulic conductivity data, SNMR is a promising technique for mapping the distribution and properties of aquifers in the near surface. Groundwater resources in the Paleozoic Bonaparte basin were mapped using a multidisciplinary, multi-physics approach. This study revealed a multi-layered hydrostratigraphy comprising of a stacked sequence of aquifers and aquitards and significant fresh groundwater resources within the basin. As part of these investigations >100 SNMR soundings were collected, inverted, and integrated with other datasets to estimate hydraulic conductivities and transmissivities for the main aquifers in the basin. Hydraulic conductivities were estimated by inserting SNMR inversion parameters T2* (decay of magnetisation) and ϕ (effective porosity) into the Schlumberger-Doll Research (SDR) equation. The ‘optimal’ formation dependent constant (C) was estimated by minimising the misfit between SNMR estimates of hydraulic conductivity and hydraulic conductivity estimates from co-located slug and pump tests. The results revealed that the median hydraulic conductivity for the three main aquifers (Keep Inlet, Kuriyippi, and Tanmurra Formations) all fall within a range of three orders of magnitude (10-1 to 102 m/day). Of these the Kuriyippi had higher transmissivities (~5 m/day) than the Keep Inlet formation (~0.8 m/day). The deeper Tanmurra formation had the highest hydraulic conductivity distribution (~30 m/day, 50th percentile), however, data were only collected at ten locations for this aquifer and, therefore, these results are unlikely to be representative of the aquifer in general, particularly at depth. This research demonstrates that SNMR is a useful technique for characterising the hydraulic properties of near-surface formations, particularly in areas where boreholes are sparse. Presented at the 2019 Australasian Groundwater Conference (AGC 2019)