Poster Sessions


P-21

Ensemble-based observation impact estimates using an observation-based verification norm

Martin Weissmann (Hans-Ertel Centre for Weather Research, LMU Munich),
Matthias Sommer (Hans-Ertel Centre for Weather Research, LMU Munich)

 
Abstract

Knowledge on the contribution of various observations and observing systems is crucial for the refinement of observing networks and data assimilation systems. This poster presents a novel approach for estimating observation impact in ensemble-based data assimilation and forecasting systems and its application to the experimental, convective-scale ensemble system of Deutscher Wetterdienst. Instead of the subsequent analysis, it uses subsequent observations for verification that are considerably more independent from the forecast. Furthermore, a peculiar property in the probability distribution of observation impact is used to define a measure for the accuracy of the impact assessment. Applying this method to a three-day test period shows that a well-defined observation impact value can be assigned to most observation groups.