Analysis of optimal and nearly optimal plans for multipurpose river projects

The climate in the Ganges-Brahmaputra basin on the Indian subcontinent is characterized by a long dry season (November to May) and a shorter intense rainy season (June to October) caused principally by the Southwest Monsoon. This has led to conflicts between India and Bangladesh, both over the scarcity of water during the period of low river flows, and with the problem of flood control during the high flows. Multipurpose river projects in the lower basin, which can provide water for irrigation, navigation and salinity control during the dry season, and help mitigate the flood waters during the rainy season, are modeled using linear programming (LP). It is shown that by generating a set of nearly optimal solutions and applying different clustering procedures to the data, a variety of distinct alternative plans can be made available to India and Bangladesh to assist in the development of water resources in the region. The model is then modified to take into account the uncertainty of the flows of the Ganges and Brahmaputra rivers. Solutions are generated and comparisons made between these new basin plans and those implemented for average flows in the rivers. The results illustrate the differences in operating policies and benefits depending upon the types of flows considered. They also indicate that certain projects could be favored in plans designed for the possibility of extreme flows.