Search

Calibration of spatially distributed models is an important issue given their over-parameterisation. Three common regionalisation methods can be distinguished: transposition, prescription and constraint. This paper proposes a strategy where the three methods are combined to provide several spatial patterns depending on the model parameters. On the one hand, insensitive and equifinal parameters are prescribed uniformly while parameters with a physical meaning are prescribed at the mesh scale. On the other hand, parameters linked with a proxy runoff signature are constrained over each sub-basin and the remaining parameters are transposed with a physio-climatic pattern constructed over the calibration sub-basins. The above tailor-made pattern regionalisation is applied at the daily time step over two large French catchments, the Loire catchment at Gien covering 35 707 km2 and the Durance catchment at Cadarache covering 11 738 km2. It is then evaluated and compared to a single regionalisation method over dozens of validation stations, treated as ungauged during the parameter regionalisation process. For that purpose, simulated and observed streamflows are compared in light of four runoff signatures: daily runoff, seasonality, flood and low flow. The results show that the tailor-made patterns succeed in significantly enhancing almost all the signatures. The enhancement appears for the least well-modelled stations and contributes a 20% improvement towards gauged modelling. It also tends to guarantee a minimum performance in an ungauged context since the minimum KGE is now 0.4 whatever runoff signature is used. [ABSTRACT FROM AUTHOR]

Model intercomparison experiments are widely used to investigate and improve hydrological model performance. However, a study based only on runoff simulation is not sufficient to discriminate between different model structures. Hence, there is a need to improve hydrological models for specific streamflow signatures (e.g., low and high flow) and multi-variable predictions (e.g., soil moisture, snow and groundwater). This study assesses the impact of model structure on flow simulation and hydrological realism using three versions of a hydrological model called MORDOR: the historical lumped structure and a revisited formulation available in both lumped and semi-distributed structures. In particular, the main goal of this paper is to investigate the relative impact of model equations and spatial discretization on flow simulation, snowpack representation and evapotranspiration estimation. Comparison of the models is based on an extensive dataset composed of 50 catchments located in French mountainous regions. The evaluation framework is founded on a multi-criterion split-sample strategy. All models were calibrated using an automatic optimization method based on an efficient genetic algorithm. The evaluation framework is enriched by the assessment of snow and evapotranspiration modeling against in situ and satellite data. The results showed that the new model formulations perform significantly better than the initial one in terms of the various streamflow signatures, snow and evapotranspiration predictions. The semi-distributed approach provides better calibration-validation performance for the snow cover area, snow water equivalent and runoff simulation, especially for nival catchments. [ABSTRACT FROM AUTHOR]

Model intercomparison experiments are widely used to investigate and improve hydrological model performances. However, a study based only on runoff simulation is not sufficient to discriminate different model structures. Hence, there is a need to improve hydrological models for specific signatures of streamflow (e.g. low and high flow) and multivariable predictions (e.g. soil moisture, snow and groundwater). This study assesses the impact of model structure on flow simulation and hydrological realism using three versions of an hydrological model called MORDOR: the historical lumped structure and a revisited formulation inflected for lumped and semi-distributed structures. In particular, the main goal of this paper is to investigate the relative impact of model equations and spatial discretization on flow simulation, snowpack representation and evapotranspiration estimate. The models comparison is based on an extensive dataset composed of 50 catchments located in French mountainous regions. The evaluation framework is founded on a multi-criteria split sample strategy. All models were calibrated using an automatic optimization method based on an efficient genetic algorithm. The evaluation framework is enriched by the assessment of snow and evapotranspiration modeling against in-situ and satellite data. The results showed that the new model formulations perform significantly better than the initial one in terms of the various streamflow signatures, snow and evapotranspiration predictions. The semi-distributed approach provides better calibration-validation performances for snow cover area, snow water equivalent and runoff simulation especially for nival catchments. [ABSTRACT FROM AUTHOR]