Variability of selected soil properties in winter wheat and native grass watersheds

The spatial and temporal transferability of saturated infiltration, surface bulk density, and surface soil composition on afield scale was investigated by a comprehensive analysis of a large data set from eight research watersheds at the ARS Grazinglands Research Station near El Reno, Oklahoma. The watersheds represented two land uses: dryland winter wheat and native warm season grasses. Variability was compared on adjacent watersheds with different land uses and on adjacent watersheds with similar land uses. Regression analysis indicated no direct relationship among the soil properties. From analysis of variance tests, grass watersheds exhibited greater saturated infiltration, lower surface bulk density, and different surface soil composition than the dryland winter wheat watersheds. Within watersheds of the two land uses, the soil properties are equivalent on all watersheds for some properties, while other properties are equivalent on only two or three of four watersheds. Saturated infiltration and surface bulk density were found to be independent of hillslope position, whereas surface soil composition was correlated with hillslope position. A limited study also indicated that saturated infiltration does not vary significantly through the growing season on the wheat watersheds. Overall, the results indicate that caution should be exercised when treating the watersheds within a single land use as equivalent units or replicates. Ultimately, this decision should be based on the specific use of the data to be collected. Study objectives that focus on the finer details of runoff production or soil-water processes may require treating each field scale watershed as a separate unit. However, if study objectives focus on seasonal water balance issues or comparisons between land uses, the field scale watersheds within each land use can be considered equivalent units.