New research demonstrates the success of digital spatial data to create a regolith-landform map with a relationship to traditional mapping.  Using data products derived by Geoscience Australia and a Geographic Information System (GIS), we created a map that had a meaningful relationship to a traditionally derived map.

Traditional regolith-landform mapping requires a large time and financial investment with ‘boots on ground’ approaches common in conjunction with interpretation of remote sensing data such as Landsat.

For mineral exploration, regolith-landform mapping is an increasingly useful tool whereby explorers can manage the regolith cover across much of South Australia for sampling programs or plan approaching the landscape to efficiently use their resources.

The method developed in this work uses spatial data and simple aggregation methods to form an objective regolith-landform map using only freely available products.  We used a 1-second Digital Elevation Model (DEM) derived by Geoscience Australia to generate Topographic Position Index (TPI) outputs at coarse (2000m) and fine (300m) radii.  These three inputs and individual gamma-ray radiometrics of uranium, potassium and thorium were all processed using an unsupervised classification in ArcGIS to produce 30 classes.  These classes were aggregated to 8 broad regolith types for further analysis.  The objective map was compared to the traditional regolith map using Mapcurves, a ‘Goodness-of-Fit’ measure evaluating spatial concordance (Hargrove et al. 2006).

The overall spatial concordance between these maps was 26.4%. Considering these maps were generated using vastly different methods, this is a positive result. Goodness-of-Fit scores between objectively mapped classes and the traditional regolith map illustrated each relationship between the maps.

This work has shown that it is possible to generate a functional regolith-landform map using spatial data products and with no prior knowledge of the region.  It will never be possible to replace traditional mapping methods, but this method can be useful when a first-pass regolith-landform map is required for mineral exploration.

This new method could be employed to enhance survey outcomes and also minimise costs of sampling unnecessary locations

This article has been published in the Quantitative Geomorphology Special Issue of Geosciences, find it here.

– Alicia (Follow me on Twitter!)