The spatial variability of saturated hydraulic conductivity was measured in situ in a clayey soil with macropores. Ksat was determined at 37 equally spaced transect sampling locations using five different-sized, in situ soil columns. Each column, starting with the largest, was constructed within the confines of the previous column. Excavation of a block of soil on four sides, with the block remaining naturally attached at the bottom, defined a volume of soil to be used in Ksat determination. The volumes used were 2.4 x 105, 1.2 x 105, 0.5 x 105, 6.28 x 103 and 884 cm3. Quantitative estimates of macroporosity were obtained from tracings of visible tubular and planar voids at the exposed upper soil surface. Measurement was also made of bulk density and percentage sand, silt, and clay. The mean (μ) and variance (σ2) of Ksat were found to be a function of the sample volume used, with the two smallest volumes producing estimated moments much larger than observed using the next two largest sizes. The 2.4 x 105-cm3-volume sample size proved to be an operationally cumbersome method and also produced extreme and probably unreliable estimates of μ and σ2. Geostatistical analysis demonstrated spatial structure to exist for the 1.2 x 105 and 0.5 x 105-cm3 sample volumes, but only random variation was observed for the other three sizes. Cross-correlation between Ksat and macrovoids and Ksat and physical properties was also demonstrated. Implications of these results for future sampling and soil survey studies are addressed.
© Williams & Wilkins 1988. All Rights Reserved.