Assessing soil corrosivity along feral-proof fencing in the Australian Arid Zone and the development of a new soil corrosivity index

Abstract

Feral-proof fencing plays a fundamental role in many wildlife conservation programs. The gradual deterioration of metal infrastructure (such as fencing) through contact with a corrosive soil environment is a complex and global issue. This paper aims to classify and map areas of low, moderate and high soil corrosivity risk along fence boundaries in the southern part of the Arid Recovery nature reserve in South Australia. This risk mapping will assist with the management of feral-proof fencing and serve as a model for understanding soil corrosion in arid environments. Maghemite and akagen ́eite were the dominant corrosion products formed on Arid Recovery fencing and are thought to be ineffective at reducing corrosion on exposed wires due to their high porosity and poor adherence to metal surfaces. On their own, soil pH, soil salinity, soil texture, soil carbonate content, soil chloride content, and soil sulfur content were all poorly correlated with the extent of fence corrosion damage observed at field sites (R2 < 0.2 and p > 0.05). The widely used American Water Works Association 10-Point Corrosion Index (C105) significantly underestimated soil corrosivity in these arid soils, with only 12% of sites correctly classified as corrosive or non-corrosive. A modified index was subsequently developed using soil chemical data, which was able to correctly classify 70% of sites. Further research is needed to accurately assess the performance of this newly developed, preliminary risk index as a reliable soil corrosivity classification method in arid environments.