Soil precompression stress assessed from linear scale stress and strain

The seemingly never-ending increase in the size and hence weight of agricultural machinery induces an urgent need for tools to evaluate the risk of subsoil compaction. The scientific community is fully aware of this as witnessed by a large number of papers addressing soil strength across soil textures, drainage conditions and other drivers of the carrying capacity of soils. The concept of soil precompression is appealing as it implies a threshold level of stress that soils may experience without plastic / permanent deformation. Much effort has been devoted to the quantification of soil precompression stress and to the potential of predicting it from soil properties. The classical procedure in determining the precompression stress from compression tests include a log-transformation of the stress imposed to the soil sample. Mathematically, the log transformation in itself introduces a bend in the strain-stress relation that does not relate to the material properties. For all procedures applied in estimating this bend for soil compression data, the estimated ‘threshold’ in the strain-log(stress) relation is an artefact or at least affected by the mathematical transformation of stress data. Despite this, efforts are still undertaken to analyze how the ‘precompression’ stress deriving from the classical procedure relates to soil properties. In the present study, we analyze data from confined, uniaxial compression tests applied to undisturbed soil cores. Our results show that it was possible to detect a local minimum of compressibility reflecting a true precompression stress. In addition, our study investigated the potential of estimating a local minimum in soil compressibility observed in studies that have measured strain at a limited number of stresses.