In our latest paper, we propose a novel framework for ensemble-based streamflow simulations. This framework develops a conceptual-data-driven approach (CDDA) that integrates a hydrological model (HM) with a data-driven model (DDM) into a hybrid modelling approach.
Thus, a HM simulates hydrological processes while a DDM simulates residuals of a HM. This framework is beneficial because it respects hydrological processes via the HM and it profits from the DDM’s ability to simulate the complex relationship between residuals and input variables.
The proposed CDDA enables exploring different DDMs to identify the most suitable model for the study at hand (catchment or model). It was tested with eight different DDMs, most promising for simulating residuals of a HM. We explored here following DDMs: Multiple Linear Regression (MLR), k Nearest Neighbours Regression (kNN), Second-Order Volterra Series Model, Artificial Neural Networks (ANN), and two variants of eXtreme Gradient Boosting (XGB) and Random Forests (RF). Based on the study in three Swiss catchments, it was found that XGB and RF provided the most promising results and these DDMs were recommended for similar applications.
The proposed CDDA framework is very flexible and can be applied to any HM and any catchment to design the most optimal DDM.
Sikorska-Senoner A.E., and Quilty J.M. (2021) A novel ensemble-based conceptual-data-driven approach for improved streamflow simulations, Environmental Modelling & Software, 143, 105094, https://doi.org/10.1016/j.envsoft.2021.105094.