Several dispersion models are available to simulate the transport processes of air pollutants and toxic substances including radionuclides in the atmosphere. Reliability of atmospheric transport models has been demonstrated in several recent cases from local to global scale, however, very few actual emission data are available to evaluate model results in real-life cases. In this study, the atmospheric dispersion of 131 I emitted to the atmosphere during an industrial process was simulated with different models, namely the WRF-Chem Eulerian online coupled model and the HYSPLIT and the RAPTOR Lagrangian models. Although only limited data of 131 I detections has been available, the accuracy of modeled plume direction could be evaluated in complex late autumn weather situations. For the studied cases, the general reliability of models has been demonstrated. However, serious uncertainties arise related to low level inversions, above all in case of an emission event on 4 November 2011, when an important wind shear caused a significant difference between simulated and real transport directions. Results underline the importance of prudent interpretation of dispersion model results and the identification of weather conditions with a potential to cause large model errors.Nowadays, several dispersion models are available to simulate the transport processes of air pollutants and toxic substances including radionuclides in the atmosphere. Reliability of atmospheric transport models has been demonstrated in several recent cases from local to global scale, however, very few actual emission data are available to evaluate model results in real-life cases. In this study, the atmospheric dispersion of 131 I emitted to the atmosphere during an industrial process was simulated with different models, namely the WRF-Chem Eulerian online coupled model and the HYSPLIT and the RAPTOR Lagrangian models. Although only limited data of 131 I detections has been available, the accuracy of modeled plume direction could be evaluated in complex late autumn weather situations. For the studied cases, the general reliability of models has been demonstrated. However, serious uncertainties arise related to low level inversions, above all in case of an emission event on 4 November 2011, when an important wind shear caused a significant difference between simulated and real transport directions. Results underline the importance of prudent interpretation of dispersion model results and the identification of weather conditions with a potential to cause large model errors. ; Other ; This work was supported by the National Research, Development and Innovation Office of Hungary (No. K116506 and No. K109109) and the New National Excellence Program of the Ministry of Human Capacities. ; Other ; {"references": ["Stein AF, Draxler RR, Rolph GD, Stunder BJB, Cohen MD, Ngan F. NOAA’s HYSPLIT Atmospheric Transport and Dispersion Modeling System. Bull Am Meteorol Soc 2015;96:2059̆201377. doi:10.1175/BAMS-D-14-00110.1", "Grell GA, Peckham SE, Schmitz R, McKeen SA, Frost G, Skamarock WC, et al. Fully coupled ̆201conlinĕ201d chemistry within the WRF model. Atmos Environ 2005;39:6957̆201375. doi:10.1016/j.atmosenv.2005.04.027.", "M̆00e9sz̆00e1ros R, Leelőssy ̆00c1, Kov̆00e1cs T, Lagzi I. Predictability of the dispersion of Fukushima-derived radionuclides and their homogenization in the atmosphere. Sci Rep 2016;6. doi:10.1038/srep19915"]}