All posts by Alain Foehn

New version 2.7.6

The RS MINERVE version 2.7.6 is now available on the CREALP website. This new version includes several improvements such as :

  • Changes in the snow model. It has been renamed from Snow-GSM to Snow-SD, which stands for Snow model with a Seasonal Degree-day factor. This annual sinusoidal variation of the degree-day factor had been implemented in the RS MINERVE version 2.4.1 in 2016. In the new version, some parameters of the snow model have been renamed, with the Reference degree-day snowmelt coefficient parameter Asn now named S. The associated parameters (AsnInt, AsnMin and AsnPh) were correspondingly also renamed (into SInt, SMin and SPh). These changes concern also the hydrological models SOCONT and GSM, which include the Snow-SD model.
  • In the GSM model, the Degree-day icemelt coefficient parameter Agl has been renamed into G, as were the associated parameters (AglInt, and AglMin into GInt and GMin).
  • The numerical computation scheme of the river models has been improved, to better preserve initial conditions when saving a model.
  • The Technical manual has been consequently modified.

Snow-modelling now based on Snow Water Equivalent

The snow-modelling part of the hydrological models integrating a snow component (Snow-GSM, GSM, SOCONT and HBV) has been improved. In former versions of RS MINERVE, the information of Snow water equivalent (SWE) was not directly available. The user only had access to two variables from which the SWE value could be computed: Hsnow, corresponding to the solid fraction of the SWE, and Theta, providing the relative water content.

From the new version 2.6, the variable Hsnow has been replaced by the variable SWE, accounting for both the liquid and the solid fraction of the snow pack:

SWE = Hsnow (solid fraction) + Wsnow (liquid fraction)

Analyses of the snow pack through an information accounting for the entire water content of the snow pack should thereby be made easier to RS MINERVE users.

Revised elevation gradient for Precipitation

When simulating in RS MINERVE, precipitation and temperature intensities are interpolated from the nearest stations using either Thiessen Polygons (nearest point) or the Shepard method, also known as the Inverse distance weighting method.

In the process of interpolation, a vertical gradient can be also used. For temperature, it is an additive term, whereas for precipitation it is a multiplicative coefficient.

By default for precipitation, it is set to 0 (no gradient).  As from the version, available as Beta since June 23rd 2017 and officially released on July 17th 2017, the unit of this gradient has been modified. The gradient of precipitation is newly expressed in [1/m], corresponding to percentage of increase/decrease per meter.

In addition, the gradient for the Potential Evapotranspiration has been also redefined in the same way.

Example : if the user wants to increase the precipitation by 20 per cent for each 1000 meters of altitude, the coefficient is equal to (20/100)/1000 = 0.0002.

The Technical manual of RS MINERVE has been consequently updated .

Improved GR4J model

The parameters introduction for the GR4J model has been slightly modified. Until version, the so-called transformed values had to be entered in RS MINERVE. As from version, released today as Beta version (available here), the real values will have to be entered.

The initial choice of using transformed values was based on users recommendations. However, using real values has the considerable advantage of having a physical meaning, for example for the store capacities in meter defined by the parameters X1 and X3. It also allows quick comparison between the parameters and the state variables.

When opening a .rsm model created with previous versions of RS MINERVE, a conversion into the new GR4J model version is proposed to automatically convert the values. Thereby, users will be able to continue working with their already existing models (Warning : once converted, no way back !).

Improved snow-melting model

Snow melting is an important component of the hydrological cycle, in particular in mountainous areas.  Modelling adequately this process is therefore important.

This led to the improvement of the snow-melting equations of the Snow-GSM, GSM and SOCONT models in RS MINERVE. Until now, the degree-day snowmelt coefficient, defining the quantity of melted snow per day as a function of temperature (mm/°C/day), was a constant factor. In the latest release of RS MINERVE, this coefficient can now vary over the year (see title image). The new parametrization is given by :

  • A reference degree-day snowmelt coefficient (Asn)
  • A degree-day snowmelt interval (AsnInt)
  • A phase shift of the sinusoidal function (AsnPh)

The first parameter defines the average degree-day snowmelt coefficient and the second parameter the interval of variation of the coefficient over the year. The third parameter defines the phase shift of the sinusoidal curve with respect to the first day of the year. As a first guess, it is recommend to use the day number of the year corresponding to the beginning of spring (80 for the Northern hemisphere, 264 for the Southern hemisphere).


Operational forecast newly based on RSMINERVE 2.0

Since July 2015, the CREALP (Centre de recherche sur l’environnement alpin) provides hydrological forecasts based on the new software version RS MINERVE 2.0.  The forecast results are used as a basis for flood prediction and flood emergency response planning for the Canton of Valais.

Three times a day, the operational system performs real-time forecasts using RS MINERVE 2.0 with command-line scripts.  Two simulation tasks are run at each time:

  • The first task, called ‘Control simulation’ simulates the last 24 hours, based on observed meteorological and measured river discharges. This task serves as control and variable initialization for the second task.
  • The second task is the ‘Forecast simulation’ which provides predicted flows for the Rhone and its tributaries, using the COSMO-7 meteorological forecast from MeteoSwiss.

Called from a DOS Terminal, each task is composed of a VBScript containing all the functions required to perform a hydrological simulation, such as launching RS MINERVE, opening the model, loading the inputs, performing the simulation and saving the results.

Documentation about the use of command-line scripts will be available with the official release of RSMinerve 2.0 in Fall 2015.


Course ‘Hydrological and hydraulic modelling with RS MINERVE’ in EPFL (Lausanne, Switzerland) – January 2015

Another session of the course ‘Hydrological and hydraulic modelling with RS MINERVE’ was held on January 28th and 29th 2015 at EPFL in Lausanne, Switzerland. The course was oriented to  end users, including engineering offices, administrations and researchers. The training was attended by 24 participants.

As for previous courses, theoretical presentations and application exercises provided the knowledge required to model hydrological basins with the help of RS MINERVE. External speakers have presented examples of hydrological applications in real projects.

In addition to the 2 days training, participants had the option to achieve at home a personal application project , proposed by the training team. Support was provided to help the participants in their work. The submitted reports were of high quality and participants showed to have well understood the functioning of RS MINERVE.

The course, taught by Javier García Hernández, Alain Foehn, Samuel Alesina and Javier Fluixá Sanmartín, from the Centre de recherche sur l’environnement alpin (CREALP) and some members of the Steering committee of Hydro10, was organized by the CREALP in collaboration with the Laboratory of Hydraulic Constructions from EPFL,  the Technical University of Valencia (Spain) and the Hydro10 Association.


New object “Structure efficiency”

The recently published version of RS MINERVE (1.1.0) offers a new object, developed by Hydro10 Association, named Structure efficiency. With a simple efficiency coefficient, it allows computing effects of discharge losses in a structure such a canal or a pipe (Output = Input * Efficiency_coefficient).

Users can now model, for example, losses in an irrigation canals network. By changing the coefficients of all Structure efficiency objects in only a few clicks, assessment of losses reduction in the network can now easily be performed.

The developments achieved in 2014 (Consumer, Hydropower, Structure efficiency, etc), trends to easily understand and improve water management to end users.

Course ‘Hydrological and hydraulic modelling with RS MINERVE’ in Lima (Peru) – May 2014

The course ‘Hydrological and hydraulic modelling with RS MINERVE’ has been held on May 6th to 8th 2014 in the Headquarters of the National Service of Meteorology and Hydrology of Peru (SENAMHI), in Lima, Peru. Theoretical and practical parts have been combined to provide the required knowledge to model hydrological basins with the help of RS MINERVE, one of the tools supported by Hydro10. About 30 participants from different Peruvian organizations have taken part with great interest and great success.

The course, taught by Javier García Hernández and Alain Foehn, from the Centre de recherche sur l’environnement alpin (CREALP) and both members of the Steering committee of Hydro10, was organized in collaboration with the SENAMHI, the CREALP, the University of Zurich, the Swiss Agency for Development and Cooperation (SDC), Meteodat, Care Perú, ProyectoGlaciares and the Ecole Polytechnique Férédale de Lausanne (EPFL).

More information (in spanish) on this webpage.