A triple-frequency approach to retrieve microphysical snowfall parameters

[1] Backscattering and extinction properties of various snow particle models are studied for three typical cloud radar frequency ranges, namely Ku band, Ka band, and W band, both in terms of their individual scattering properties as well as averaged over size distributions. Models studied include soft spheres, randomly oriented pristine nonspherical particles and complex aggregates, as well as horizontally aligned spheroids. It is shown that the concurrent use of Ku/Ka band and Ka/W band dual wavelength ratios (DWR) allows for a separation of different snow particle habits. It is further shown that triple-frequency approaches constrain the slope parameter of exponential size distributions more tightly than conventional single DWR approaches can. Uncertainties introduced by unknown mass-size relations for different snow particle habits remain a challenge when mass-related quantities are to be derived. Attenuation by snow, especially at W band, is found to potentially alter these results, albeit moderately, without affecting the general conclusions. Sensitivity studies performed with respect to cutoffs in the simulated size distribution highlight potential benefits of including larger particles in future scattering databases.