Field traffic may reduce the amount of air-filled pores and cavities in the soil thus affecting a large range of physical soil properties and processes, such as infiltration, soil water flow and water retention. Furthermore, soil compaction may increase the mechanical strength of the soil and thereby impede root growth. The objective of this research was to test the hypotheses that: (1) the degree of soil displacement during field traffic depends largely on the soil water content, and (2) the depth to which the soil is displaced during field traffic can be predicted on the basis of the soil precompression stress and calculated soil stresses. In 1999, field measurements were carried out on a Swedish swelling/shrinking clay loam of stresses and vertical soil displacement during traffic with wheel loads of 2, 3, 5 and 7 Mg at soil water contents of between 11 and 35% (w/w). This was combined with determinations of soil precompression stress at the time of the traffic and predictions of the soil compaction with the soil compaction model SOCOMO. Vertical soil displacement increased with increased axle load. In May, the soil precompression stress was approximately 100 kPa at 0.3, 0.5 and 0.7 m depth. In August and September, the soil precompression stress at 0.3, 0.5 and 0.7 m depth was 550–1245 kPa. However, when traffic with a wheel load of 7 Mg was applied, the soil displacements at 0.5 m depth were several times larger in August and September than in May, and even more at 0.7 m depth. An implication of the results is that the precompression stress does not always provide a good indication of the risk for subsoil compaction. A practical consequence is that subsoil compaction in some soils may occur even when the soil is very dry. The SOCOMO model predicted the soil displacement relatively well when the soil precompression stress was low. However, for all other wheeling treatments, the model failed to predict that any soil compaction would occur, even at high axle loads. The measured soil stresses were generally higher than the stresses calculated with the SOCOMO model. Neither the application of a parabolic surface load distribution nor an increased concentration factor could account for this difference. This was probably because the stress distribution in a very dry and strongly structured soil is different from the stress distribution in more homogeneous soils.
Read full abstract