The stochastic dual dynamic programming (SDDP) algorithm introduced by Pereira and Pinto in 1991 has sparked essential research in the context of water resources management, mainly due to its ability to address large-scale multistage stochastic problems. This paper aims to provide a tutorial-type review of 32 years of research since the publication of the SDDP algorithm. A systematic academic literature search identified 174 scientific papers on water resource management published in 96 different journals. A bibliometric analysis is conducted to identify the main methods used to tackle this type of problem and to determine recent and future research trends. Our analysis reveals that stochastic dynamic programming, which was initially the most used approach, has now been replaced by multistage stochastic programming. Risk-averse and robust approaches are also gaining strength in recent years due to uncertainty related to climate change. Water inflows have been the main source of uncertainty considered in the literature by far, followed by, e.g., electricity demand, electricity prices, fuel costs, and renewable energy availability. In addition, as computational capacity continues to increase, aspects of nonlinearities, disaggregated networks, and different water management strategies are increasingly considered to make modeling more realistic. This work suggests there is still a need for tractable stochastic optimization models for large-scale power and water systems that deal with multiple uncertainty sources and nonlinearity approximations.