We study the reservoir management problem in hydropower plants, and propose operating policies to maximize the average annual energy generation or the average annual revenue. Under revenue maximization, our policies allow for short-term electricity price variations to be incorporated into the long-term plan. First, we provide a detailed review of hydropower plant operation, focusing on implicit stochastic optimization approaches and integration of energy price variations in reservoir management. Then, we explain non-linear programming models that we developed for obtaining operating policies with different characteristics. We evaluate and compare the operating policies through a case study. Policies characterized by dynamic end-of-month storage levels are shown to perform much better than the policy with an optimal static end-of-month storage level, and it has been further shown that the dynamic policies perform quite close to the theoretical upper bound. Finally, we show that maximizing the average annual energy and maximizing the average annual revenue objectives yield considerably different operating policies and using one policy in place of the other may result in significant loss of benefit or resource.