| dc.description.abstract | The main objective of this thesis was to investigate the error in the wind speed and wind direction predicted by the AROME MetCoOp model for the time period after the Chernobyl accident, using radiosonde data with and without adjusting for horizontal displacement. The second objective was to examine specific radiosonde soundings to investigate the error in wind direction along the movement of the plume over the first four days, using radiosonde observations with horizontal drift and discussing the significance of these errors for the transportation of the radioactivity.
To do this, the radiosonde observations were adjusted for horizontal drift by using methods explained in Laroche and Sarrazin (2013), and it was applied to the analysis of the model wind predictions, so that it could be compared to the analysis without adjustment. Then the model data was interpolated for the data to have the same position in space and time as the radiosonde observations. Then the mean absolute error and mean error were found for each timestep and for different height bins. Before this there was a scatter plot analysis of the mean absolute error and removal of outliers. To investigate the error in selected radiosonde stations, polar plots were presented to visualize the difference in predicted and observed wind direction, giving a picture of the difference in the observed and predicted path of the plume.
Among other findings, this thesis concluded that the horizontal displacement of radiosondes, when comparing radiosonde observations with the AROME model, was not necessary to include in the analysis seeing that there were very little difference between the analysis with adjustment for horizontal displacement of the radiosonde observations and the analysis without. It was further observed that there was a larger error below 1000 meters than above, and an under prediction of the rotation of the wind direction from the surface of the earth to the free atmosphere. Therefore this thesis also concluded that depending on the vertical position of the radioactivity there could be more or less deviations between the model predictions and the actual transportation. Further, this thesis also came to the conclusion, based on the errors in wind direction, that radioactivity above 1000 meters during the first few days could be transported more north than predicted, and the radioactivity below 1000 meters could be transported more towards the west than predicted. | |
