Landscape-scale exploratory radiometric mapping using proximal soil sensing

Abstract

The soil's gamma radiometric signal is known to be a crucial indicator of the heterogeneity of soil properties, most likely directly related to the mineralogy and geochemistry of the parent material. However, fine-resolution radiometric data sourced from geophysical surveys are often not available. In our study area in the Lower Hunter Valley, Australia, only coarse-scale (based on 2km line spacing) aerial radiometric maps are available. An alternative solution could be to implement a ‘smart’ ground-based gamma-ray survey when detailed radiometric information is required on larger extent. The work presented here therefore details a vehicle-borne gamma-ray soil survey conducted in the Lower Hunter Valley to investigate if a proximal soil sensing device with a footprint much smaller than airborne gamma-ray spectrometers can be employed for regional extent mapping, and still capture the soil's fine-scale heterogeneity. To do so we designed an algorithm for the delineation of irregular transects (drive-lines) based on ancillary environmental information, intended to be used specifically for Wide-Ranging Exploratory Surveys (WIRES) at the regional scale. The WIRES algorithm creates transects from randomised starting points that terminate at boundary points; the path is defined with directional constraints that collectively ensure that the environmental variability within the domain of study is accessed and sampled. The design was delineated for the entire 220km2 study area. In this exploratory work we surveyed a sub-catchment of the area (≈15km2) to trial the feasibility of WIRES. We produced radiometric maps based on a regression-kriging approach. An external validation of the maps showed good results. We also validated the maps produced with existing detailed radiometric maps across a landholding (≈140ha) that is situated within the sub-area investigated in this study. Both surveys were comparable, which gives us useful insights into understanding the scaling properties of the geophysical properties of soil across environments.