Field measurements of icing parameters and their application towards ice load maps: An overview

Ice load maps are used for the purposes of planning and design of infrastructure in ice-prone regions. They are obtained based on microphysical output from long-term numerical weather prediction model simulations. A common microphysics scheme used in such simulations is the Thompson “aerosol-aware” parameterization scheme implemented in the Weather Research Forecasting (WRF) model. The generation of ice load maps relies on a few key assumptions regarding icing parameters, primarily, that the Median Volume Diameter (MVD) can be accurately calculated using gamma distribution from the values of Cloud Droplet Number Concentrations (CDNC) and Liquid Water Content (LWC) from WRF. The validation of this assumption is hampered due to difficulties of measuring these parameters in field. Recent technological advancements allow for these measurements. Such a measurement campaign was carried out, and its results were implemented in an analytical scheme, mimicking that of an ice load map calculation scheme. The measurement results show that the assumed droplet concentration values are overestimated in WRF. The analytical calculations results show that the generalized gamma distributions in Thompson “aerosol-aware” scheme is adequate for representing icing conditions when supplied proper input data. The WRF results are deemed useful for further use in the generation of ice load maps, as-is, however, there are now open questions about how the newly observed characteristics and connections between CDNC, LWC and MVD can impact the icing rates in particular at such exposed maritime sites. A re-evaluation of the measured icing data on a finer time resolution scale and de-coupled from WRF model output may be warranted to improve the obtained results. However, modifications to the droplet activation parameterization of the Thompson “aerosol-aware” scheme and/or the background CCN concentrations would be required to reproduce the measured droplet distribution properties.