Multi-Hydro is a fully distributed physically based model developed at the Ecole des Ponts
ParisTech. It is build on the four open source software developed separately and widely used in
the scientific world.
With its modular structure, Multi-Hydro can be easily adapted to the need of each case study
. As it is GIS based and it don't need to be calibrated, Multi-Hydro is easily transportable to a
case study to another one. The GIS data are quickly assimilated with the help of MH-AssimTool,
an open source sofware developed at the Ecole des Ponts ParisTech.
The Multi-Hydro Surface Module (MHSM) models the surface runoff and relies on TREX, which was
developed by HydroQual Incorporate and the Colorado State University (Velleux et al., 2011). It
uses spatially distributed data (hydrological network, land use, type of soil, elevation,
overland storage depth and the initial conditions) to model the behaviour of the catchment.
TREX website !
The Multi-Hydro Ground Module (MHGM) models sub-surface processes and relies on the VS2DT
model, which was developed by the U.S. Geological Survey (Lappala et al., 1987). It uses finite
differences to compute the pressure head, the total head, the moisture contents and/or the
saturation and the mass balance, with the help of the law of conservation of fluid mass and a non-linear form of the Darcy equation. It takes into account the structure and the composition of the soil, the amount of water entering in the system, the temperature and the initial conditions.
VS2DT website !
The Multi-Hydro Drainage Module (MHDM) models flows in sewer systems and relies on SWMM, which is developed by the United States Environmental Protection Agency and the Camp Dresser & McKee incorporation (Rossman, 2010). The model basically relies on 1D Saint-Venant equations, which are used to dynamically simulate sewer flow. The sewer system is described in details (pipe, node, shape, length, slope, initial head, Manning’s coefficient, etc).
SWMM website !
The Multi-Hydro Rainfall Module (MHRM) enables to downscale rainfall data. Indeed the usual
C-band radar resolution of 1km*1km*5min does not enable to fully take advantage of the high
resolution distributed hydrological model (10m). The space-time downscaling is performed with
the help of universal multifractals cascades (Shertzer and Lovejoy, 1987). More details about
the multifractal analysis and downscaling of this process can be found in Gires et al. (2011).