Hydrological models

Hydrological models

What models do we develop? An increase in the time step (equivalent to look at the functioning of the catchment with a coarser temporal resolution) must be accompanied by a simplification of the model structure.

Model structures have been specifically developed for the daily, monthly and yearly time steps (see our publications for more details).

Modélisation

The models are the following:

  • GR4J : GR model with 4 parameters for the daily time step
  • GR2M : GR model with 2 parameters for the monthly time step
  • GR1A : GR model with 1 parameter for the yearly time step

Perrin et al. (2007) present a summary of these three models. Our research and development studies also comprise the establishment of model structures tailored to specific needs of operational services (e.g. flood forecasting or drought forecasting), as well as routines to better simulate key hydrological processes in the catchment:

In our studies, we search to extensively test our models on a large sample of catchments and to develop robust methods for model calibration and validation.

Crash test for a standardized evaluation of hydrological models. Hydrol. Earth. Syst. Sci 13,1757-1764 - © Andréassian et al.(2009)

In this folder

Brief history of works that led to the development of the GR models

Bandeau GRP WEBGR

GRP is a a hydrological model developed for short-range flood forecasting (from a few hours to a couple of days), at hourly time step.

The figure below presents the GR4J model (in French, modèle du Génie Rural à 4 paramètres Journalier). More recent versions of the model exist: with five parametres (GR5J; Le Moine, 2008) and six parametres (GR6J; Pushpalatha, 2013). They were developed to provide a better simulation of low flows.

The last version of the GR2M model was developed by Mouelhi (2003).

There are several versions of the annual GR model.

The evaluation of 27 potential evapotranspiration (PET) formulas for rainfall-runoff modelling applications (Oudin, 2004; Oudin et al., 2005) led to a simple and efficient PET formula that allows obtaining better results than the other formulas tested, in terms of simulation of streamflows.

The CemaNeige snow module was implemented to improve discharge modelling at the outlet of snow-influenced catchments.