Real-time prediction of soil infiltration characteristics for the management of furrow irrigation

Abstract

The spatial and temporal variations commonly found in the infiltration characteristic for surface-irrigated fields are a major physical constraint to achieve higher irrigation application efficiencies. Substantial work has been directed towards developing methods to estimate the infiltration characteristics of soil from irrigation advance data. However, none of the existing methods are entirely suitable for use in real-time control. The greatest limitation is that they are data intensive. A new method that uses a model infiltration curve (MIC) is proposed. In this method a scaling process is used to reduce the amount of data required to predict the infiltration characteristics for each furrow and each irrigation event for a whole field. Data from 44 furrow irrigation events from two different fields were used to evaluate the proposed method. Infiltration characteristics calculated using the proposed method were compared to values calculated from the full advance data using the INFILT computer model. The infiltration curves calculated by the proposed method were of similar shape to the INFILT curves and gave similar values for cumulative infiltration up to the irrigation advance time for each furrow. More importantly the statistical properties of the two sets of infiltration characteristics were similar. This suggests that they would return equivalent estimates of irrigation performance for the two fields and that the proposed method could be suitable for use in real-time control.