Know more

The overall aim of the DRHYM project is to explore the potential of data-driven hybrid models for hydrological applications and to demonstrate their added value in real-life case studies. By hybrid, we mean models developed by integrating machine learning algorithms with conceptual hydrological models in a novel common framework.

The main paradigm of the project is to consider the catchment as a source of information to identify improved hybrid model structures with wider applicability and robustness in time and space. This will be done by putting the search for improved model structures in hydrology in the framework of artificial intelligence (AI) techniques and large data sets. The objective is to produce models with improved predictive skills for practical applications of a wide range of end-users. The DRHYM project aims to answer the following specific questions:

1. How can AI-based techniques be used to develop efficient and flexible hybrid models?
2. Can we improve the versatility, generality and robustness of hydrological models?
3. To which extent do data-based hybrid models need data?

In the DRHYM project, we will demonstrate the added value of the hybrid models for practical applications that need to (i) extrapolate hydrological models to data-scarce conditions and (ii) rely on robust simulations in non-stationary climate conditions.