Submitted to: Vadose Zone Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/2/2005
Publication Date: 11/1/2005
Citation: Wuest, S.B. 2005. Bias in ponded infiltration estimates due to sample volume and shape. Vadose Zone Journal. 4:1183-1190.
Interpretive Summary: Estimates of saturated and unsaturated water flow in soil are important for predictions of infiltration, runoff, and solute transport. These critical estimates are usually based on measurements made using relatively small samples. Previous research indicated that ponded infiltration rate estimates might be influenced by the volume or cross-sectional area of the measurement. Our study determined that the diameter of cylinder used for ponded infiltration measurement in the field has a direct effect on the rate estimates obtained. Larger cross-sectional areas caused very significant increases in infiltration rate estimates. Further testing indicated that the sample-size bias is due to lateral flow of water, which is restricted to a greater degree in small cylinders than in larger cylinders. These findings are important because efforts to model water and fluid flow through soil are currently limited by a lack of understanding as to how various soils differ from theoretical constructs of fluid flow through porous media. Simply recognizing that measurements of saturated flow need to be relatively large in order to estimate true water infiltration at a field scale should improve the accuracy of some existing models.
Technical Abstract: Estimates of saturated and unsaturated water flow in soil are important for predictions of infiltration, runoff, and solute transport. Previous research indicates that ponded infiltration estimates are influenced by the volume or cross-sectional area of the measurement. Our study compared infiltration measurements made using 20-, 30-, and 45-cm diameter cylinders driven 25 cm deep into field plots under diverse agricultural management practices. Mean infiltration rate increased from 50, to 81, to 95 mm h^-1 as diameter increased. Standard deviation and range also increased with diameter. All three diameters produced log-normal data distributions. These results indicate increasing sample size was not equivalent to the pooling of smaller samples, which would have produced the same mean but with a lower variance. Follow-up experiments with a double-ring configuration or a divider placed in the center of a 45-cm cylinder indicated that restricted lateral flow of water between preferential flow pathways was responsible for the reduction in infiltration. A pulse of dye introduced 10 min before removing the ponded water showed an extensive network of preferential flow pathways in all but the slowest infiltration situations. The pathways were not associated with visible macropores. Air permeability was well correlated with water infiltration. Careful consideration should be given to the dimensions of samples used to estimate saturated and possibly unsaturated flow, given the apparent influence of lateral preferential flow paths.