Technical Abstract: Soil hardpans found in many of the Southeastern USA soils reduce crop yields by restricting the root growth. Site-specific soil compaction management to alleviate this problem requires determination of the spatial variability and mapping of soil hardpans. The objective of this study was to determine the spatial variability of soil hardpan as influenced by soil moisture. Geo-referenced soil cone index measurements were taken in 200 grid cells (10 m X 10 m grid cell size) on Pacolet sandy loam soil (Fine, kaolinitic, thermic Typic Kanhapludults) in Auburn, AL (USA) on June 29, 2004 and August 25, 2004 representing wet and dry soil measurement dates. Core samples were also taken in 5 cm depth increments up to a depth of 65 cm for soil moisture and bulk density determinations. Statistical and geostatistical methods were used for the data analysis. In the 0-35 cm depth, the soil moisture had dried significantly by August 25, 2004 (Dry) as compared to the soil moisture on June 29, 2004 (Wet). An isotropic spherical semivariogram model best fit the semivariances of the peak cone index for wet and dry soil conditions. Soil drying increased the peak cone index and the maximum semivariance value. Small but statistically significant differences were also observed on the predicted depth to the peak cone index as the soil dried in the 0-35 cm depth. In the dry soil condition, the semivariances of the predicted depth to the peak cone index were nearly constant over the separation distances suggesting that the depth to the hardpan did not exhibit substantial spatial dependence.