New triple-wavelength radar measurements to test scattering models for aggregate snowflakes

Radar is a powerful tool to probe the microphysics of ice clouds and snowfall, but our ability to retrieve snowflake size spectra from single- or dual-wavelength reflectivity measurements is limited by our understanding of the scattering properties of individual snowflakes (which in turn depends on their geometry)

Recently, Kneifel et al 2011 have suggested that adding a measurement at a third wavelength may allow us to constrain the scattering properties of snowflakes. However very little triple-wavelength data is available to test these ideas.

Results are presented from the Chilbolton Triple Wavelength Snowflake Experiment at the Chilbolton Observatory in southern England during 2014/15. Data in deep stratiform ice clouds were collected with colocated 3, 35, and 94 GHz radars. We find that the triple-wavelength signatures of natural aggregate snowflakes are inconsistent with simple spherical and spheroidal models, and that such models may lead to errors of several dB at 94GHz. In contrast, we find that the data is well described by the ensemble response of a large number of random realisations of fractal aggregates generated by a physical aggregation model.

We also show that when the snowflakes approach ~ 1cm in size the dual-wavelength ratio for 35-94GHz "saturates" at a fixed value between 8-9dB, at which point the dual-wavelength ratio ceases to contain any information on particle size (the 3-35GHz dual wavelength ratio in contrast continues to increase). Theoretical considerations show that this saturation is consistent with fractal geometry, and the specific value allows the fractal dimension to be deduced (~2 in our case).