Validation of Microphysical Snow Models Using In Situ, Multifrequency, and Dual‐Polarization Radar Measurements in Finland

As complex forward models for snow have become common in radar‐based retrievals, there is a demand to validate these models in different environments. In this study, we perform a qualitative, general validation for nine different snow models that have been published and are available to users. The chosen models span a variety of different snow types, such as aggregates, rimed aggregates, melted aggregates, graupel, and single crystals, mainly because these particles are commonly observed in the Finnish climate. Fitted power law formulas for mass, fall velocity, aspect ratio, and area ratio are compared between the models and 5‐year winter measurements in the Hyytiälä forestry field station in Finland. We also compare the backscattering properties of the models to triple‐frequency dual‐polarization radar measurements during the Biogenic Aerosols Effects on Clouds and Climate campaign in 2014. We find that the denser models, regardless of the exact shapes, fit the in situ measurements best due to the prevalence of rime in the falling snow. However, when comparing also to the triple‐frequency radar measurements at X, Ka, and W bands, and the linear depolarization ratio at Ka band, the physical snow models fit overall better than the empirical ones. Key Points Denser snow models fit the in situ measurements overall better due to the presence of rime in Finnish snowfall Physical aggregate models are able to cover a larger area of triple-frequency and dual-polarization radar features than empirical models The physical aggregate models and the fractal model were overall best fitting for both the in situ and the radar measurements