| Field experiments were performed to quantify the changes in infiltration rate due to surging using blocked furrow infiltrometer. Six soils were tested using combinations of on-time and off-time duration of surge and continuous flow. The soils had clay percentages ranging from 7 to 28 percent. The on-times were 30, 60, and 90 minutes for the first two soils. Off-times were 30, 60, 120, 180, and 960 minutes.;Due to the variability in infiltration rate, reduction of infiltration rate (RIR) was used to standardize the reduction of infiltration rate between surge one and surge two.;Furthermore, statistical analyses were performed to determine the effects of the management variables on the reduction of infiltration rate. Tests were performed on the reduction infiltration rate data using one-way analysis of variance, ANOVA, where the independent factors were the on-time, off-time, and clay content. The reductions of infiltration rate under surge irrigation were significantly higher than those for continuous flow at the 95 percent level for all six soils.;The results of this study indicate that the RIR increased with the increase of off-time up to the optimum off-time, then the reduction decreases. Furthermore, off-time had a significant effect on RIR at the 95 percent level for all soils, while on-time had a significant effect at the 95 percent level on only two soils: no. 4 and soil no. 6).;A regression model was derived from the field data to predict surge irrigation RIR using management variables and clay percentage. The predictions of the regression model were then compared to the field data. The regression model explained 61 percent of the variance of the reduction of infiltration rate. The regression model had high correlation with the field data.;Killen (1988) developed an analytical model by combining the Green-Ampt equation and a simple non-hysteretic redistribution model (Gardner et al., 1970) to predict infiltration rate under surge irrigation. The results of infiltration rate predicted by the Killen model were subjected to same standardization that was used for the field data. The results predicted by the Killen model were then compared to field data. |
