MAPPING FIELD-SCALE PHYSICAL-PROPERTIES OF SOIL WITH ELECTRICAL-RESISTIVITY

The spatial variability of physical properties that significantly infl uence the fate of water and solute in soils needs a large number of me asurements to be quantified. Surface electrical resistivity techniques could be used as a simple and practical method to determine this spat ial variability. Electrical sounding and profiling measurements were t aken on a small agricultural field (30 by 60 m) under two different so il conditions (dry and wet conditions). The soil profile is composed o f three layers: a highly permeable sandy loam (alluvial terrace) overl ying a gravelly sandy till that covers a friable sandy to silty shale. The soil physical properties (grain size distribution, porosity, hydr aulic conductivity, bulk density, and organic matter content) of the u ppermost layer were measured in the laboratory on undisturbed soil cor es taken at three different depths on a 6 by 15 m grid in the field. C orrelations were established between these parameters and the electric al conductivity. The best correlations were between the electrical con ductivity and the sand, silt, clay, and organic matter contents. Their correlation coefficients, r, were, respectively; 0.64, 0.53, 0.64, an d 0.65 for the dry conditions and 0.54, 0.45, 0.53, and 0.52 for the w et conditions. No relation was established between the electrical cond uctivity and the porosity, the bulk density, or the hydraulic conducti vity. The correlation coefficients were, respectively, 0.04, 0.16, and 0.10 for the dry conditions and 0.14, 0.12, and 0.14 for the wet cond itions. Electrical conductivity seems to be more influenced by the soi l texture, i.e., by the electrical properties of the soil constituents , than by the structure, i.e., the water-related properties. Besides, the two sets of electrical resistivities obtained in dry and wet condi tions are not significantly different, as shown by the regression betw een them (slope = 0.92, r = 0.71) and by the isoresistivity maps. This study seems to indicate that the electrical method could be used to e valuate the spatial variability of some soil properties when their var iability is sufficiently large, i.e., when the investigation scale or the level of contrast is large enough.