The effect of spatial resolution on the in-cloud atmospheric icing conditions in numerical weather model at the Fargernes mountain, Norway

Abstract

Atmospheric icing poses significant risks to infrastructure, aviation, and the energy section. Numerical weather models, as the Weather Research and Forecasting model (WRF), can be used to describe the atmospheric conditions relevant for atmospheric icing. In this study, we will focus on the Fagernes mountain meteorological icing measurement site where the WRF model is set up using ERA5 input data, Thompsons microphysics scheme to describe the different hydrometeors, and the Yonsei University (YSU) planetary boundary scheme with increasing spatial resolutions from 9, 3 and 1 km resolution. The final high-resolution model is using a Large Eddy Simulation (LES) for planetary boundary layer option in WRF with 91 m horizontal resolution in the model, in which a 10 m digital elevation model of Norway is used as model input.

Using supercooled liquid water content, we have shown that the increased resolution from 9 to 1 km clearly changes the atmospheric conditions in the numerical model at the Fagernes mountain icing rig site. The main reason for this change seems to be that higher resolution models provides a better representation of the true terrain. Since the icing rig is located on a mountain top, the model height of the site does increase as the model spatial resolution increases.

Introducing the high-resolution LES model, both the mountain height the surrounding terrain is clearly closer to the real terrain at the measurement site. The LES model provides very good results for studying single or short time icing events and allows for a better understanding of the local terrain effects when in comes to atmospheric icing. The increased computational cost of the LES model makes it difficult use for larger areas and/or for long time simulations.