The paper deals with the problem of optimizing the multipurpose use of seasonal storage pools in the power system with a significant share of hydropower. The fundamental goal is to achieve maximum profit from different ways of using water, taking into account all possible limitations. The main difference between the proposed method and the usual method of pool management is in the treatment of other water users. Other users can continue to be modeled as constraints (eg water supply to the population) but also as equal participants in the realization of profits. The modeling of other water users and their objective functions is, at least for now, quite simple compared to the modeling of hydropower plants and their profits in the power system. All input variables such as inflows into pools, intermediate inflows, demands of electricity consumers, etc. are modeled deterministically. In the proposed method, the optimization problem for S basins is broken down into S subproblems in which one basin is optimized while the discharges of the other basins are kept temporarily constant. The solution of the subproblem, i.e. the determination of the optimal flow distribution on one reservoir, can be found by a suitable optimization procedure for solving nonlinear optimization problems with constraints. Therefore, a long-term water use policy or a solution to a global objective function results from successively solving a series of subproblems described by local objective functions. By decomposing the objective function into two levels, we make the optimization problem simpler and solvable with proper management of the entire procedure and the use of a suitable optimization method for frequently solving the local objective function. The local objective function contains fewer variables, has a simpler structure, and is therefore easier to solve. Thanks to the decomposition of the problem, each phase of the calculation can also include some less influential parameters, which were previously neglected. The possibly longer duration of the entire budget is an insignificant factor when solving problems of this type. The proposed method is applicable for long-term planning in the conditions of the open market of production and sale of electricity due to the possibility of calculating the electricity balance with different electricity prices in certain time periods. The proposed method was tested on several examples.