Multi-model data fusion as a tool for PUB: example in a Swedish mesoscale catchment

Abstract

Abstract. Post-processing the output of different rainfallrunoff models allows one to pool strengths of each model to produce more reliable predictions. As a new approach in the frame of the "Prediction in Ungauged Basins" initiative, this study investigates the geographical transferability of different parameter sets and data-fusion methods which were applied to 5 different rainfall-runoff models for a low-land catchment in Central Sweden. After usual calibration, we adopted a proxy-basin validation approach between two similar but non-nested sub-catchments in order to simulate ungauged conditions.Many model combinations outperformed the best single model predictions with improvements of efficiencies from 0.70 for the best single model predictions to 0.77 for the best ensemble predictions. However no "best" data-fusion method could be determined as similar performances were obtained with different merging schemes. In general, poorer model performance, i.e. lower efficiency, was less likely to occur for ensembles which included more individual models.

Improving streamflow estimation in ungauged basins using a multi-modelling approach

“…In a very few studies, multiple structures of hydrological models have been combined for prediction in ungauged basins (e.g. Goswami et al 2007, Exbrayat et al 2011. Exbrayat et al (2011) combined the outputs of five rainfall-runoff models using a data-fusion and weighting approach for daily runoff prediction at an ungauged basin.…”

Application of the CNOP‐P Ensemble Prediction (CNOP‐PEP) Method in Evapotranspiration Forecasting Over the Tibetan Plateau to Model Parameter Uncertainties

“…Since the proposal of the HEPEX, some efforts have been made to obtain improved hydrological predictions. For example, Exbrayat et al (2011) employed the multimodel ensembles (MMEs) method to obtain runoff predictions using 5 different rainfall-runoff models in a low land catchment in Central Sweden. The authors pointed out that the ensemble runoff predictions obtained with the MME method were better than the single-model predictions.…”

<i>Preface</i> Towards practical applications in ensemble hydro-meteorological forecasting

“…Based on 29 catchments in France, Velázquez et al (2011) compare 16 different lumped hydrological model structures driven by weather forecasts from the European centre for medium-range weather forecasts (ECMWF) and evaluate the uncertainties associated with model structures and determine if performance improves when multiple hydro-models are implemented. Exbrayat et al (2011) study the geographical transferability of different parameter sets and data-fusion methods applied to 5 different rainfall-runoff models for a catchment in Central Sweden. L. evaluate the performance of a statistical post-processor for imperfect hydrologic model forecasts.…”