Digital soil mapping enables informed decision-making to conserve soils within protected areas

Abstract

Protected areas are regarded as pristine land, but often they require rehabilitation and effective management to prevent increased land degradation. Soil management should be based on soil maps, which are difficult to create in protected areas due to their large size, restricted accessibility, limited available soil data and low budgets for such projects. The objective of this paper is to showcase a novel hybrid expert knowledge and machine learning digital soil mapping (DSM) method to map soils covering large areas with limited accessibility and available soil data, and on a small budget. The study is situated at Benfontein, a 9 900 ha protected area in South Africa. Soil landscape rules were used to determine virtual soil observation locations which were added to the training dataset used by a machine learning algorithm to create an acceptable soil associations map (validation Kappa = 0.69). Soil properties and interpreted soil indices were assigned to each soil association at 0.1, 0.5 and 0.9 percentile levels, to indicate the range of properties at an 80% certainty. Results show that Benfontein has large carbon sequestration potential, the soils are relatively stable against water erosion, and off-road driving should be prohibited on approximately half of the area. The approach of percentile mapping of soil property ranges at high confidence levels optimises limited data. The hybrid DSM method is viable for creating useful soil maps in data-scarce environments to inform management decisions in the unique settings of protected areas.