INFLUENCE OF SPATIALLY-VARIABLE SOIL HYDRAULIC-PROPERTIES ON PREDICTIONS OF WATER-STRESS

Models are often used to predict drainage system effects on crop production. Variability of soil hydraulic properties on predictions made by many models have not been evaluated. This study evaluated several methods for predicting water transport properties and corn (Zea mays L.) stress as influenced by variable soil hydraulic properties in a field of Portsmouth sandy loam (Typic Umbraquults). Upflux, drainage volume, and infiltration parameters as functions of water table depth were predicted using hydraulic conductivity and soil water retention functions for three soil horizons measured at 150 locations in a field. Crops stress due to both deficient and excess soil water conditions and relative crop yield were estimated using DRAINMOD, a water management stimulation model, for three selected methods of averaging soil property inputs. Small differences existed among the three approaches for the 30-yr average relative corn yield. Large differences in relative corn yield occurred in dry years indicating that the variability of the soil properties was important to consider in predicting crop stress during relatively dry years. More information for the field soil-drainage response was obtained using the individual locations method which allowed soil property input to vary from location to location at each of the 150 points in the field. However, the field averages approach is more practical because fewer data are required to perform the necessary computations and only a 3% difference in the 30-yr relative yield resulted between the individual location and field average methods.