 Hybrid hydroclimatic forecasting systems combine data-driven statistical and machine learning methods with physics-based models to produce more accurate predictions of meteorological and hydrologic variables and events. These systems have been gaining popularity due to advancements in weather and climate prediction systems, increased computing power, and access to large datasets. The goal of these systems is to reduce the effects of bias in dynamical models and take advantage of the strengths of machine learning algorithms. Additionally, hybrid forecasting systems can be used to combine multiple sources of predictability with varying time horizons, which can help to increase accuracy and reduce uncertainty. This article was authored by L. J. Slater, L. Arnold, M. A. Boucher, and others.