Modeling Optimal Water Management for Reservoir Based Irrigation Projects

The double cropping of rice in the Muda Irrigation Project depends very much on the volume of water stored in the Muda and Pedu reservoirs. The most important problem affecting the project includes constraints related to poor water management. However, the shortage of reservoir water still remains the most severe constraint on the establishment of stable double cropping of rice. Thus, the main purpose of this study is to develop an optimization model and a solution strategy to solve the water resources of the project in a computationally satisfactory manner. In this study, a water balance model was developed and the performance of a project was evaluated. The water balance components were modeled, without incorporating any model calibration parameters. The model results were compared with observed data satisfactorily. The overall project efficiency for the main and off-seasons were also obtained. A reservoir simulation model was developed and the model storage capacities were compared with the observed storage capacities satisfactorily. The Markov process with periodicity in hydrologic data was applied to generate monthly streamflows. The generated storage capacities were found to simulate with the observed storage capacities satisfactorily. Three different cases of anticipated future monthly storage were envisaged to assess the risk for predicted monthly storage capacities in 1998-2002 and their probabilities of occurrences were computed. An optimization model was also developed to solve the water resources management of a large project in a computationally satisfactory manner. Twelve different scenarios were analyzed to test the performance of the project and their consequences were illustrated. The optimal reservoir storage, optimal irrigation demand, and optimal reservoir release (i.e., optimal reservoir operating policy) were computed. The optimal mean (1987-1997) model total water requirements for the dry and wet seasons were also computed and the optimal contributions by rainfall, reservoir, uncontrolled river flow, and recycled water were determined. The mean water balance components results for different months were stored in GIS data bases, analyzed, and displayed as the monthly crop water requirements maps. Finally, it can be concluded that the integration of the water balance model together with the models for reservoir simulation, efficiency, hydrologic forecasting, optimization, and GIS holds much promise in the analysis of optimal allocation of water resources of a project.

PDF FK_1999_43_A.pdf Download (1MB)

Actions (login required)

View Item