In Situ Subsoil Stress-Strain Behavior in Relation to Soil Precompression Stress Thomas Keller, 1,2 Johan Arvidsson, 2 Per Schjønning, 3 Mathieu Lamande ´, 3 Matthias Stettler, 4 and Peter Weisskopf 1 Abstract: Soil compaction negatively influences many important soil functions, including crop growth. Compaction occurs when the applied stress, R, overcomes the soil strength. Soil strength in relation to com- paction is typically expressed by the soil precompression stress, R pc . Deformation is assumed to be elastic and reversible as long as R e R pc . This work examined soil stress-strain behavior as measured in situ during wheeling experiments and related it to the stress-strain behavior and R pc measured on soil cores in uniaxial compression tests in the laboratory. The data analyzed were from a large number of wheeling experiments carried out in Sweden and Denmark on soils with a wide range of texture. Contradicting the concept of precompression stress, we observed residual strain, ? res , at R e R pc . These observations were supported by stress-strain data measured in uniaxial compression tests, which likewise showed ? res 9 0 at R e R pc . Residual strain was observed in the field when R exceeded approximately 40 kPa, and when the ratio R/R pc exceeded roughly 0.1, although ? res was very small at R/R pc G 0.5. These values were similar to those obtained on confined uniaxial compression curves. On the basis of our findings, we question the use of R pc as a measure of soil strength and call for a reevaluation of the precompression stress concept. Key words: Soil compaction, soil strength, soil deformation, wheeling experiment. (Soil Sci 2012;177: 00Y00) S oil compaction negatively influences many important soil functions, including crop growth. It is therefore officially recognized as one of the main threats to soil, for example, by the Thematic Strategy for Soil Protection of the European Union. Because the regeneration of soil functions is a slow process, especially in the subsoil (Ha ˚kansson and Reeder, 1994), it is generally accepted that soil compaction should be avoided (Alakukku et al., 2003). Compaction, that is, volumetric deformation of soil (which is expressed as an increase in bulk density or a decrease in total porosity), occurs when the applied stress, R, overcomes the soil strength. Soil strength in relation to compaction (compression) is typically expressed by the precompression stress, R pc (Casagrande, 1936; Lebert and Horn, 1991). Deformation is assumed to be elastic, reversible, as long as R e R pc , and plastic, irreversible, if R 9 R pc (Hartge and Horn, 1984). This has been referred to as the ‘‘precompression stress concept’’ (Horn and Fleige, 2003, 2009; Lebert et al., 2007). Hence, soil compaction could be avoided by limiting the stresses exerted by agricultural machinery to below R pc (Alakukku et al., 2003; Horn and Fleige, 2009). Therefore, R pc is an important property in soil compaction mod- eling (De ´fossez and Richard, 2002) and is used in recommenda- tions and regulations for the protection of soil against compaction (Lebert et al., 2007). Although R pc is widely used as a measure of compressive strength in agricultural soil mechanics, the validity of the pre- compression stress concept has been little tested against field data, that is, against in situ measured stress and deformation during wheeling. The reason may be that such in situ measure- ments are laborious and difficult to standardize. During the last decade, a number of wheeling experiments have been carried out by Keller and Arvidsson and coauthors (Arvidsson et al., 2002; Keller et al., 2002; Keller and Arvidsson, 2004). In the present study, we collated and analyzed these data. The objective was to study soil stress-strain behavior as mea- sured in situ during wheeling experiments and to relate it to the stress-strain behavior and soil precompression stress as mea- sured on soil cores in uniaxial compression tests in the laboratory. MATERIALS AND METHODS Wheeling experiments were carried out in Sweden and Denmark in the period 2000 to 2004. Some of the data used here have been included in previous publications with the focus on stress propagation (Table 1). The texture of the soils studied ranged from sandy loam to clay (Table 1). Wheeling experiments were performed with towed trailers, wheeled tractors, and sugar beet harvesters, with wheel loads in the range 17 to 103 kN (Table 1). All experiments were carried out on arable soil in autumn (after harvest, before autumn tillage) or spring (before spring tillage). Sampling of undisturbed soil cores for determi- nation of soil mechanical properties in the laboratory was carried out in combination with the wheeling experiments. In Situ Measurements of Vertical Stress and Displacement Vertical soil stress and vertical displacement were measured by installing probes into the soil horizontally from a dug pit that was approximately 1.5 m long, 1 m wide, and 1 m deep, with the walls stabilized with wooden boards (Fig. 1A). The probes were installed through drilled holes that were stabilized by inserting a steel tube with the same diameter (58 mm) as the hole. For each wheeling pass, three probes were installed, typically at 0.3-, 0.5-, and 0.7-m depth. The distance between the pit wall and the probe head was approximately 1.1 m (Fig. 1A). Stress was measured by a load cell (DS Europe Series BC 302, Milan, Italy) with a diameter of 17.5 mm (Fig. 1B). Determination of TECHNICAL ARTICLE Soil Science & Volume 177, Number 8, August 2012 www.soilsci.com 1 1 Agroscope Reckenholz-Ta ¨nikon Research Station ART, Department of Natural Resources and Agriculture, Zurich, Switzerland. 2 Swedish University of Agricultural Sciences, Department of Soil & Envi- ronment, Uppsala, Sweden. 3 Aarhus University, Department of Agroecology, Research Centre Foulum, Tjele, Denmark. 4 Bern University of Applied Sciences, School of Agriculture, Forest and Food Sciences, Zollikofen, Switzerland. Address for correspondence: Thomas Keller, Agroscope Research Station ART, Reckenholzstasse 191, 8046 Zu ¨rich, Switzerland; E-mail: thomas.keller@ art.admin.ch Financial Disclosures/Conflicts of Interest: None reported. Received December 23, 2011. Accepted for publication May 31, 2012. Copyright * 2012 by Lippincott Williams & Wilkins ISSN: 0038-075X DOI: 10.1097/SS.0b013e318262554e Copyright © 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.