ESTIMATION OF INFILTRATION AND HYDRAULIC RESISTANCE IN FURROW IRRIGATION, WITH INFILTRATION DEPENDENT ON FLOW DEPTH

Estimation of infiltration and hydraulic resistance model parameters from furrow irrigation evaluation data was investigated. A semi-physical, flow-depth dependent furrow infiltration model was used for the analysis. Macropore infiltration was modeled empirically as a constant volume of water per unit area that is absorbed instantaneously. The estimated infiltration parameters were the saturated hydraulic conductivity and the macropore constant. Hydraulic resistance was modeled with the Manning equation; thus, the estimated resistance parameter was the Manning coefficient. The estimation procedure uses volume balance calculations and unsteady flow simulation. Estimation of the flow-depth dependent infiltration parameters requires first the estimation of an empirical infiltration function, via volume balance, and of the resistance parameter, via unsteady flow simulation. Simulated flow depth hydrographs as a function of distance are then used as inputs for a second set of volume balance calculations with the semi-physical infiltration model. This procedure takes advantage of the fact that similar surface flow conditions (advance and recession times, flow depths, and runoff rate) can be predicted with different infiltration functions. The procedure was tested with two irrigation data sets, each consisting of three furrows. With both data sets, the semi-physical infiltration model fitted the measured volume balance data as well as empirical infiltration models. For each group of furrows, estimates of the hydraulic conductivity were of comparable magnitude and were consistent with published values for the particular soil texture. Likewise, estimates of the macroporosity parameter were consistent for each group of furrows. Estimates of the Manning coefficient suggest very uniform hydraulic resistance. Finally, results show that the effect of flow-depth dependent infiltration on irrigation distribution uniformity depends on soil and hydraulic conditions. For one of the data sets, distribution uniformity computed with the proposed infiltration model was only slightly different from the uniformity computed assuming that infiltration depends only on opportunity time, because of the large macropore flow contribution and the shallow flow conditions.