Chris Turnadge

Biography

Chris’ research involves characterising various aspects of regional-scale groundwater flow systems. Locations of recent assessments have included the Eyre Peninsula in South Australia, the Pilbara region of Western Australia, and the Roper River catchment in Northern Territory. Chris uses a wide range of analytical and numerical tools for groundwater flow system characterisation. These approaches include the interpretation of hydraulic, hydrochemical, and isotopic data, including hydraulic tests and environmental tracers.

Building on research developed during the Deep Earth Imaging Future Science Platform, Chris is collaborating with Flinders University (Australia) and the University of Georgia (USA) to develop novel methods for estimating subsurface hydraulic properties via the performance of sinusoidal hydraulic tests. These tests can be more insightful, more efficient, and provide more robust results than traditional hydraulic tests. This research was awarded the 2022 Innovation and Technology Award by IAH Australia and is currently being commercialised through a partnership with an Australian company, Kremford Pty Ltd, to produce the Limax sinusoidal slug test system.

In collaboration with the University of Newcastle (Australia) and the University of Georgia (USA), Chris is developing methods for the estimation of subsurface properties from groundwater responses to natural tidal influences. These include ocean, atmospheric, and Earth body tides, which can be interpreted using established digital signal processing methods. These approaches provide cost-effective alternatives to traditional hydraulic tests and are accessible via the https://groundwater.app website and the HydroGeoSines package.

Chris is collaborating with the University of Colorado Boulder (USA) to develop and apply adjoint state methods for more efficient means of calculating sensitivities of subsurface flow and transport models. Although first developed over 50 years ago, the uptake of adjoint state methods has historically been limited by poor communication and the absence of ready implementations. These methods can provide insights into the relative value of various observed data types, as well as enabling more highly parameterised models while reducing inversion run times.