Journal of Irrigation and Drainage Engineering

The ability of evolutionary neural networks (ENN) to model reference evapotranspiration (${\mathrm{ET}}_0$) was investigated in this study. The daily climatic data, solar radiation, air temperature, relative humidity, and wind speed of three stations in central California, Windsor, Oakville, and Santa Rosa, were used as inputs to the ENN models to estimate ${\mathrm{ET}}_0$ obtained using the FAO-56 Penman-Monteith equation. In the first part of the study, a comparison was made between the estimates provided by the ENN and those of the following empirical models: the California Irrigation Management System, Penman, Hargreaves, modified Hargreaves, and Ritchie methods. Root-mean-squared error, coefficient of efficiency, and correlation coefficient statistics were used as comparing criteria for the evaluation of the models’ accuracies. The ENN performed better than the empirical models. In the second part of the study, the ENN results were compared with those of the conventional artificial neural networks (ANN). The comparison results revealed that the ENN models were superior to ANN in modeling the ${\mathrm{ET}}_0$ process.

A procedure for evaluating water and energy use in pressurized irrigation networks was developed. Performance indicators were derived from the International Program for Technology and Research in Irrigation and Drainage (IPTRID) and the Institute for Diversification and Energy Savings (IDAE) in Spain and applied to ten representative irrigation districts with on-demand pressurized networks during the 2006–2007 irrigation season. The results confirm the high energy requirements needed for operating these irrigation schemes. To apply an average depth of $2,589{\mathrm{m}}^3/\mathrm{ha}$, the energy required was estimated to be $1,000\mathrm{kW}\mathrm{h}/\mathrm{ha}$. Power requirements per unit of irrigated area were $1.56\mathrm{kW}/\mathrm{ha}$ and the pumping energy (PE) efficiency was 58%.

Small reservoirs known as tanks are constructed in the watersheds of arid, semiarid, and subhumid regions of India to provide supplementary or protective irrigation to crops during dry spells of the monsoon season or full irrigation during the postmonsoon season. The stored water in tanks or recharged groundwater is used for this irrigation. Several models have previously been developed to design the capacity of individual tanks. However, for optimum utilization of water generated in a watershed to meet the demands for irrigation and for downstream release, it is necessary to design the tanks together in terms of their number, locations, and capacities. A comprehensive methodology for this is presented using stream points, i.e., possible tank locations on the main stream(s) in the watershed. Tank strategies (combinations of numbers of tanks, their locations at stream points, and tank types) are then generated for the identified stream points. Subsequently, fields in the watershed are assigned to the catchment and the command of different tanks of a specified tank strategy. Simulation of field, tank, and groundwater balance is then carried out on a daily basis, from which optimum tank dimensions are obtained for a specified tank strategy. The optimum tank strategy and corresponding optimum tank dimensions are obtained by investigating all the possible tank strategies.

The SOFTANK model optimally designs the watershed-based tank system by simulating field, tank, and groundwater balances. We applied this model to a small watershed consisting of six tanks (small reservoirs) in the semiarid region of India. We evaluated the existing tank system in this watershed and compared it to a one-tank system. Results showed that one tank at the outlet of the watershed would have been more beneficial [with benefit-cost (BC) ratio of 1.80] than the existing six-tank system (with BC ratio of 1.71). Finally, we performed the analysis for obtaining the optimal tank system for the watershed, and we found that the tanks for irrigation purposes are not economical for the small watershed. The groundwater source was enough for irrigation, so any additional investment in the tanks would be uneconomical. The results demonstrate the importance of the watershed-based tank system approach to design.

Sharp open-channel bends are commonly encountered in hydraulic engineering design. Disturbances such as secondary flows and flow separation caused by the bend may persist for considerable distances in the downstream channel. A simple way of reducing these disturbances is through the insertion of vertical vanes in the bend section. A laser Doppler anemometry (LDA) unit was used to measure the three-dimensional mean and turbulent velocity components of flow in an experimental rectangular open-channel bend. Flow characteristics of the bend with no vanes are compared with those of bends having one or three vertical vanes. The size of the flow separation zone at the inner wall of the bend was determined from dye visualization data and confirmed with mean streamwise velocity data. Results show that the vertical vanes are effective in considerably reducing flow separation, intensity of secondary flows, and turbulence energy in the downstream channel. Furthermore, energy loss for bends with vanes is slightly less than for the no-vane case.

Weir head-discharge relationships are typically described using the discharge coefficient-dependent standard weir equation. The submerged weir (tailwater exceeds the weir crest elevation) head-discharge relationship can vary from the free-flow head-discharge relationship, particularly at high submergence levels. The accuracy associated with predicting the upstream head or discharge, corresponding to submerged weir flow conditions, is dependent upon the accuracy with which a representative submerged discharge coefficient can be determined. A submerged ogee crest weir discharge coefficient predictive method developed by the U.S. Bureau of Reclamation (USBR) is reviewed and its predictive accuracy compared to laboratory-scale submerged ogee crest weir experimental data associated with a wide range of submerged flow conditions for nine different ogee crest weir geometries. Discussion is presented in an effort to partially explain the relatively poor correlation between the USBR method and the experimental data set. Alternative submerged discharge coefficient relationships are also presented.