Modelling atmospheric icing: A comparison between icing calculated with measured meteorological data and NWP data

In this case study several periods of measured active icing events have been compared with calculated active icing. Icing events from three winter seasons in Sweden have been selected as a base for studying how well a commonly used method to calculate icing can capture icing events, as well as to try out a new approach. Another purpose is to enlighten parties interested in atmospheric icing of the challenges of calculating icing. For those with little knowledge of Numerical Weather Prediction (NWP) models it can be hard to understand why there are differences between two models, or even the same model using different physical parameterisation schemes. In this study data from two NWP models have been used to calculate icing, as well as being compared with a new approach using meteorological measurements to calculate icing. Using meteorological measurements to calculate icing is an interesting alternative as it is computationally less expensive than using a NWP model to generate the necessary data. All tested methods could at times capture the dynamics of the measured icing, though there are on occasion large differences between the different calculated icing and also between calculations and measurements. This highlights the importance of having an understanding of NWP models when studying atmospheric icing; there can be a large spread between different models, and also when using the same model with different schemes. The results show that calculated icing is comparable to measured icing and that the methodologies investigated are viable options for calculating atmospheric icing. (C) 2015 Elsevier B.V. All rights reserved.