1Institute of Meteorology and Climate Research, Karlsruhe Institute of Technology (KIT), Herrmann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany
2Deutscher Wetterdienst (DWD), Frankfurter Straße 135, 63067 Offenbach, Germany
Received: 28 Jan 2013 – Discussion started: 22 May 2013
Abstract. An extended version of the German operational weather forecast model was used to simulate the ash dispersion during the eruption of the Eyjafjallajökull. As an operational forecast was launched every 6 hours, a time-lagged ensemble was obtained. Sensitivity runs show the ability of the model to simulate thin ash layers when an increased vertical resolution is used. Calibration of the model results with measured data allows for a quantitative forecast of the ash concentration. After this calibration an independent comparison of the simulated number concentration of 3 μm particles and observations at Hohenpeißenberg gives a correlation coefficient of 0.79. However, this agreement could only be reached after additional modifications of the emissions. Based on the time lagged ensemble the conditional probability of violation of a certain threshold is calculated. Improving the ensemble technique used in our study such probabilities could become valuable information for the forecasters advising the organizations responsible for the closing of the airspace.
Revised: 12 Mar 2014 – Accepted: 23 Jun 2014 – Published: 07 Aug 2014
Vogel, H., Förstner, J., Vogel, B., Hanisch, T., Mühr, B., Schättler, U., and Schad, T.: Time-lagged ensemble simulations of the dispersion of the Eyjafjallajökull plume over Europe with COSMO-ART, Atmos. Chem. Phys., 14, 7837-7845, doi:10.5194/acp-14-7837-2014, 2014.