The EFAS post-processing methodology is based on Model Conditional Processor (MCP), it is used to correct the discharge forecasts for specific locations, so they become predictors of future observed discharge value. This is done through the formulation of a probabilistic model for the future observations given the available observations, simulations and forecasts. 

Figure - Representation of the joint probability distribution of observations and simulations, from Biondi, Daniela & Todini, Ezio. (2018). Comparing Hydrological Postprocessors Including Ensemble Predictions Into Full Predictive Probability Distribution of Streamflow. Water Resources Research. 10.1029/2017WR022432. https://doi.org/10.1029/2017WR022432

In EFAS, the post-processing is composed of two parts; the calibration (off-line), and the real-time part:

Calibration (off-line)

• Estimation of a joint probability distribution of observations and simulations, where the simulation for each station comes from LISFLOOD long term run. 24-hourly data from observations and simulations are used at the moment (6-hourly data will be used in the future)
• Generation of the probabilistic model parameters

Post-processing (real-time)

• It's performed for each station where calibration and real-time discharge data are available 
• It uses the probabilistic model parameters, observations from last 40 days and ENS forecasts (ECMWF-ENS, ECMWF-HRES, DWD-HRES, COSMO-LEPS)
• It provides post-processed results on 24-hourly time steps (6-hourly steps will be available in the future)
• There are more than 1000 pp station available in EFAS

EFAS post-processed forecasts are available for the relevant stations as an additional graphic shown in the pop-up window of the 'Reporting Points' layer.

Two examples are shown below, for the stations ID 334 - Bellinzona, Ticino (Switzerland), and ID 173 - Federaun, Gail (Austria).  

Figure - Real time hydrographs for stations Bellinzona (left), and station Federaun (right).