Water level measurements, discharge measurements (gaugings) and rating curves constitute an indirect way of estimating the discharge in rivers whose level of uncertainty deserves to be described. Given the growing importance of environmental questions and resulting regulatory requirements and all innovative technologies that are emerging in the field of hydrometry (LSPIV, SVR…), it is necessary to know the uncertainty of such estimates.

However, most modellers presently ignore the often high uncertainty associated with streamflow data in the calibration of hydrological models and use these data as if they were totally accurate. The calibration of such models could change drastically if uncertainty were taken into account. This could have an impact on flood forecasting and the design and operation of infrastructures that directly concern the safety of life and property. Major errors and damage can occur if the estimation of streamflow uncertainty is neglected, that is why the quantification of streamflow data uncertainties is an extremely important issue in hydrology that needs to be accounted for.

Thus, the aim of the proposed session is to underline this idea by giving hydrometers the opportunity to demonstrate how big the uncertainty around constructed streamflow data series can be by showing the uncertainty of each step of their construction (water level measurements, gaugings, tracing of the rating curves…). By presenting new technologies, sensors, ways to build the rating curves ans case studies taking into account the idea of uncertainty, the goal of such a session is to show that streamflow data series are sometimes far to be accurate.