SPATIAL VARIABILITY OF SOIL WATER RETENTION FUNCTIONS IN A SILT LOAM SOIL

Soil water characteristic curves are a prerequisite for quantifying the field soil water balance and predicting water flow in unsaturated soils. The spatial variation of water retention in the root zone influences water availability for plants, evaporation, and fluxes of water and solutes through soils. The purpose of this study was to determine the ability of a popular model for the soil water retention function to describe the spatial variability of measured retention data and to investigate the application of a water content scaling theory to reduce the apparent spatial variation of soil water retention. Using a combination of Tempe cells and 1.5-MPa pressure plate extractors, we measured soil water retention at six pressure heads. In total, 281 undisturbed soil core samples were taken from the Ap horizon (0 to 17-cm depth increments) along an 80-m transect on a bare silt loam soil at 0.30-cm intervals. Sample statistics were calculated to identify outliers and erroneous data. A four-parameter retention model (θ s , θ r , α, n) was fitted to the data, and water content scale factors were also calculated. The soil water retention model was found to be extremely flexible in fitting the measured data. The parameters in the retention model showed a structured variance with a range of influence between 12 and 30. The number of parameters needed to characterize the field variability was 912 for the retention model. Scaling theory applied to the water retention data significantly reduced the apparent spatial variability. One scale factor also showed a structured variance, indicating a spatial correlation distance of greater than 30 m. Using the Akaike information criterion, we found that scaling theory could adequately represent the spatial variation in water retention with only 460 parameters. Sampling, calibration and/or experimental errors were thought to account for more than 50% of the total variability