In August 2005, a intense flashflood occurs on the Vorz catchment affecting the village of Saint-Agnès. This event highlighed the difficulties to forecast the hydrometeorological conditions in mountain areas where they are extremely variable in space and time (spatially and temporally) and frequently poored monitored. From this observation a project was funded to implement an original meteorological monitoring system on the catchment, in order to observe the natural and hydrologic phenomena to better understand them and to build methods and tools for their modeling. After two years of monitoring, the first results showed that the network implement allows to obtain informations on hydrometeorological process at high spatial and temporal resolution. In spite of the installation in a harsh mountain environment (access, cold, energy,...) a very good reliability, and a lot of perspectives of transpositon on other catchments have been point up for low investment costs.The originality of the network is to achieved a multi-sampling on a lot of hydrometeorological parameters (rain, temprature, snow, insulation,...), with spatial (10 to 50 meters) and temporal (hourly or less) resolution to performed a hydrological modeling at different scale both for the water ressource management (long term) or flashflood prevention (short term). The Sensors use in the network constitute a complementary and indivisible set of monitoring system: iButtion (air and soil temperature), rain gauge, totalizer, cameras. The development of an innovative sensor for automatic cartography of the snow cover (SnoDEC) from terrestrail photographies was achived during this work. This sensor allows to quantify spatial and temporal heterogeneity of snow cover evolution on the catchment, with images taken at regular time steps (5 to 7 frames per day). This heterogenity is essential for understand and modelling the hydrology considering the strong snow persistence (5 to 10 months). The dense network set up on the catchment enable us to collect a large database and implement different interpolation techniques on hydrometeorological process on the catchment. Thus, accurate maps of temperatures and rain are created with a daily or hourly timestep. Furthermore, the SnoDEC sensor will permit to analyse and quantify the spatial and temporal heterogeneity (elevation, aspect, velocity of melting,...) of the snow cover. From the database, we will better understand the hydrological mechanisms occuring on the site, and we will build the first ideas and method for the future modelisation. In the same time, the available data will be combinate in order to highlight phenomena very difficult to measure (rain/snow limit, thermical inversion,...) and that will be use in the future to constraint accurately the snow and hydrologic models. Because of the different parameters monitored and the use of imagery, the network is able to measure variables from many field of study (glacier dynamic, vegetative cycle,...). Thereby, with its interdisciplinary approach the network think to implement a monitoring system at low cost in destination of the actors of study and research in mountain.
