Simulation of peak-demand hydrographs in pressurized irrigation delivery systems using a deterministic–stochastic combined model. Part I: model development

Abstract

A deterministic–stochastic combined model named HydroGEN was developed, as described in a companion paper (Part I: Model development), to enable the simulation of demanded daily volumes and hourly flow rates during peak periods in pressurized irrigation delivery networks. The model was applied to a pilot large-scale irrigation system located in southern Italy for calibration and for testing its reliability in analyzing the operation of large-scale pressurized delivery systems through the simulated flow configurations. Daily input data on rainfall, temperature, solar radiation, wind speed and relative humidity were gathered from a meteorological station located within the study area, whereas information on local irrigation management practices were collected through interviews with farmers and from extension specialists. The model was tested at different management levels, from district to sector and hydrants. The model testing was supported by the use of high-resolution remote-sensing imagery acquired on a single overpass date in 2006 and then classified and recoded following a ground-truthing campaign conducted during the same year. Simulations were performed to identify the 10-day peak-demand period and to generate the hydrographs of daily volumes and of hourly flow rates. Results from the different simulations were compared with historical datasets of irrigation volumes and discharges recorded during the 2008 and 2009 seasons at the upstream end of the irrigation network under study, at a sector level during the 2007 season and at selected delivery hydrants during the 2005 season. Some discrepancies between simulated and recorded data were noted that can be related to small errors in estimating crop and soil parameters, application efficiency at field level, as well as to large variability in irrigation management practices followed by local farmers. Overall, the results from testing showed that the model is capable of forecasting with good accuracy the timing of peak-demand periods, the irrigation volumes demanded during the season, as well as the hydrographs of daily volumes and hourly flow rates withdrawn by farmers during these peak-demand periods, especially when it is applied to large multi-cropped command areas.