Growing volumes of hydrological and remote-sensing data make data-driven modeling an increasingly powerful complement to physically based approaches. This theme applies machine learning, genetic programming, and surrogate models to prediction and decision support.
Applications include rainfall–runoff and streamflow forecasting, water-quality modeling, and fast surrogates that make large optimization studies tractable.
- Forecasting with machine-learning and evolutionary models
- Surrogate modeling for optimization
- Hybrid data-driven and physics-based methods