Evaluation of time domain reflectometry (TDR) for estimating furrow infiltration

TDR was used to estimate furrow infiltration, which is a key component in furrow irrigation system design and management. Furrow irrigation experiments were conducted on bare and cropped fields consisting of three 40 m long parabolic shaped furrows spaced at 0.8 m on a slope of 0.5%. The centre furrow was taken as the study furrow and the other two provided a buffer to the centre furrow. Altogether, 22 irrigations were conducted during 2004 and 2005 with inflow rates ranging from 0.1 to 0.7 l s⁻¹. TDR probes were installed vertically around the centre furrow at four locations 0.5 (S₁), 13 (S₂), 26 (S₃) and 39.5 m (S₄) from the inlet end. The S₁ and S₃ locations had four TDR probes installed at 0.15, 0.30, 0.45 and 0.60 m depths whereas the S₂ and S₄ locations had two probes each at 0.15 and 0.30 m depths. Soil moisture data collected at 5-min intervals were used to determine the average soil moisture content of the field. The change in moisture content was used to estimate the furrow infiltration which was compared with that measured using an inflow–outflow (IO) method. The performance of the TDR method was studied by calculating the absolute prediction error (APE), root mean square error (RMSE) and index of agreement (I _a). It was found that the TDR-method estimated furrow infiltration well for higher inflow rates and during the initial stages of irrigation. APE decreased and I _a increased with increase in flow rate for both bare and cropped conditions. The APE and RMSE were found to be larger for a cropped field than the bare field when irrigated at the same inflow rate. The accuracy of the TDR-method for estimating total infiltration was improved by using the average field moisture content of 30 or 45 min after the recession phase ceased. These results indicate that TDR can be used to estimate in situ infiltration under furrow irrigation.