TIME-SERIES MODELING OF CHANNEL TRANSMISSION LOSSES

The transmission losses through a canal network depend on a number of physical, operational, and climatic factors. Some of these factors can be quantified but most of them are not easy to be incorporated into p recise calculations. Due to climatic and other uncertainties, predicti on of the reliable amount of losses in irrigation projects cannot be d one confidently. Because of these uncertainties, the water loss series observed over a period of time can be treated as a stochastic process that can be modeled by time series techniques. In the present study, the transmission losses series observed at the Imperial Irrigation Dis trict, California, USA, is modeled as an autoregressive-integrated-mov ing average (ARIMA) process. The ARIMA models are found to be plausibl e and appropriate models for such series, These are simple to develop and provide dependable forecasts. Reliable forecasts of canal transmis sion losses are useful in the efficient management of irrigation proje cts.