Deriving Snowfall Microphysical Properties and Testing Connection with Triple-Frequency Radar Observation

The capacity of deriving snowfall properties from remote sensing measurements relies on the availability of accurate microphysical and scattering models of snowflakes. In recent years, large efforts have been made in improving our understanding of snowflake growth processes, microstructure, single scattering properties and fall behavior. Despite its importance in fostering the development of snowfall retrieval methods, the consistency among those various physical models of snowfall properties has not been extensively tested yet.

In the present study, the connection between ice particle microphysical properties and triple-frequency radars signatures is investigated by using observations from X/Ka/W- band cloud radars, video particle imager, 2D-video disdrometer and weighting precipitation gauge, obtained during the Biogenic Aerosols Effects on Clouds and Climate field campaign that took place in Finland in 2014. The purpose of this investigation is to verify the feasibility of a common modeling framework able to simultaneously match the observation obtained by a collection of ground instruments and a coincident triple-frequency radar measurement.

This conceptual framework is mainly divided into two parts: first, the microphysical properties of falling snowflakes are derived by retrieving mass-size relationships in form of a power-law that are consistent with ground observations and X-band radar reflectivity; then, these microphysical properties are used as an input for accurate single particle scattering models to calculate the expected radar reflectivity at higher frequencies.

The microphysical properties of falling ice particles are quantified by their ensemble mean density of snowfall and snowflake mass-size relation. It is shown that various definitions of the ensemble mean density are sensitive to distinct moments of the particle size distribution and may yield significantly different results. The magnitude of this discrepancies is related with the value of the exponential term of power-law expression of ice particle mass-size relation. Based on the observed differences between definitions of ensemble mean snow densities, a method to retrieve snowflake mass-size relation from coinciding measurements of effective radar reflectivity factor, liquid water equivalent accumulation and ice particle volume flux is developed.

The proposed method yields ice particle mass-dimensional relations that are consistent with the X-band radar, weighing gauge and particle video imager observations. The retrieved relations are also in agreement with ones independently retrieved by applying general hydrodynamic theory to 2D-video disdrometer data.

To test how well a relation between snow microphysical and scattering properties is maintained across multiple radar frequencies, Ka- and W- band radar reflectivity observations were calculated using soft spheroid and complex ice particle DDA models. These calculations are compared to Ka- and W- band cloud radar measurements and it is shown that complex ice particle models and corresponding discrete dipole approximation computations represent a better link between physical and scattering properties of ice particles with respect to the soft spheroid model.

The results of this study shows that, under some assumptions, it is possible to simulate consistent radar reflectivities across X,Ka and W bands using the DDA snowflake scattering model and a coincident retrieval of snowfall microphysical characteristics.