Abstract
Statistical post-processing of short-term hydrological ensemble forecasts using the application of the dressing method
Highlights
Simplification of reality in prediction models, inaccurate input data and other sources of uncertainty lead to predictions that always, more or less, differ from observation
The post-processing method dressing with a dynamically compiled run-time error model is a functional tool for adjusting ensemble hydrological forecasts which are based only on ensemble precipitation forecasts
Methods increase the success of hydrological ensemble predictions by including uncertainty of hydrological modelling
Summary
Simplification of reality in prediction models, inaccurate input data and other sources of uncertainty lead to predictions that always, more or less, differ from observation. The presented method includes the uncertainty of hydrological modelling into the calculation of the ensemble hydrological forecast It is primarily intended for the improvement of probabilistic forecasts based exclusively on precipitation variants. The method was tested in order to increase the success of operational ensemble predictions which serve as an irreplaceable source of information for river navigation in the Elbe and for the management of water reservoirs with regard to optimizing electricity production and minimizing the impact of drought It is applied as a post-processing procedure, which means adjusting the hydrological forecast after its output from the hydrological model. The dressing method combines the already created hydrological ensemble forecast, which is based on the probabilistic prediction of precipitation, with the statistical distribution of deviations of hydrological modelling, and achieves a comprehensive description of the entire uncertainty of the hydrological forecast
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
