Soil nitrate-nitrogen (NO3-N) data arranged on a three-dimensional grid were analyzed to compare tillage effect on the spatial distribution of residual NO3-N in the soil profile of agricultural plots drained by subsurface tiles. A three-dimensional median-based resistant (to outlier(s)) approach was developed to polish the spatially located data on soil NO3-N affected by directional trends (nonstationarity in the mean) in the three major directions (row, column, and depth) and along the horizontal diagonal directions of the grid. Effect of preferential or nonpreferential path of transport of NO3-N in the vertical direction defined as sample hole effect was also removed to make the data trend-free across holes. Composite three-dimensional semivariogram models (along horizontal and vertical directions) were used to describe the spatial structure of residual soil nitrate distribution. Two plots in the same field, one under each tillage system (conventional tillage and no tillage), were studied. In each plot, soil samples were collected at five depths (30, 60, 90, 120, and 150 cm) from 35 sites (holes) arranged on a 7¿5 regular grid of 7.6¿7.6 m. In the conventional tillage plot, residual NO3-N concentrations decreased gradually to a depth of 90 cm and increased beyond this depth. The coefficient of variation, however, became gradually smaller, showing more uniform distribution for greater depths. In the no-tillage plot, trends were similar to those in the conventional tillage system, but were spatially more stable across the profile. Structural analyses indicated that under conventional tillage, the semivariogram of residual soil nitrate distribution was linar in the horizontal and vertical directions. In contrast, the semivariograms for no-tillage showed nugget-type behavior, indicating a lack of spatial structure in the residual soil nitrate.