Spatially constrained inversion for quasi 3D modelling of airborne electromagnetic data - an application for environmental assessment in the Lower Murray Region of South Australia

We present an application of spatially constrained inversion (SCI) of SkyTEM (airborne electromagnetic) data for de. ning spatial patterns of salinisation in the Bookpurnong irrigation area located in the lower Murray Basin of South Australia. SCI uses Delaunay triangulation to set 3D constraints between neighbouring soundings, taking advantage of the spatial coherency that may be present in the dataset. Conductivity information for individual soundings is linked through the spatial constraints, from well determined parameters to locally poorly determined parameters. For the survey presented here, SCI generated maps detail the spatial variability of foodplain salinisation, the extent of foodplain sediments influenced by lateral recharge and flushing along stretches of the Murray River, and the variable quality of groundwater in deeper semi-confined aquifers of the Murray Group. Available borehole and other ancillary information, such as vegetation density and health patterns, match the observed conductivity variations seen in the SCI results, even at the very near surface (approximate to 2m depth). The SCI provides more accurate and spatially consistent results compared with those from single site inversions. They are also more uniform and detailed than maps obtained with single point Layered Earth Inversions or a laterally constrained inversion. In this example, the SCI provided reliable quasi 3D modelling, that confirmed and improved the hydrogeological knowledge of the area, indicating that the technique would have application with helicopter electromagnetic data in similar settings throughout the lower Murray Basin of Australia.