Ori et al 2014

Ori, D., T. Maestri, R. Rizzi, D. Cimini, M. Montopoli, and F. S. Marzano (2014), Scattering properties of modeled complex snowflakes and mixed-phase particles at microwave and millimeter frequencies, J. Geophys. Res. Atmos., 119, 9931–9947, doi:10.1002/2014JD021616
Link to the database
Davide Ori
T. Maestri, R. Rizzi, D. Cimini, M. Montopoli, F. S. Marzano
Ori et al 2014
The aggregation scheme in four steps


Types of particles

aggregates, mixed-phase


10 to 157 GHz

Scattering properties

Radar backscattering cross section, Total extinction cross section, Full phase function , Total absorption cross section, Total scattering cross section


0.1 to 15 millimeters


230 to 270 K

Random orientations?


Shapefiles included?


Scattering method(s)


Scattering method details

A-DDA interdipole spacing 0.02 millimeters


A microphysically based algorithm (named Snow Aggregation and Melting (SAM)) that models snowflakes composed of a collection of hexagonal columns by simulating a random aggregation process is presented. SAM combines together pristine columns with multiple dimensions to derive complex aggregates constrained to size-mass relationship obtained by data collected from in situ measurements. The model also simulates the melting processes occurring for environmental temperatures above 0°C and thus define the mixed-phase particles structure. The single-scattering properties of the modeled snowflakes (dry and mixed phased) are computed by using a discrete dipole approximation (DDA) algorithm which allows to model irregularly shaped targets. In case of mixed-phased particles, realistic radiative properties are obtained by assuming snow aggregates with a 10% of melted fraction. The single-scattering properties are compared with those calculated through Mie theory together with Maxwell-Garnett effective medium approximation using both a homogeneous sphere and a layered-sphere models. The results show that for large-size parameters there are significant differences between the radiative properties calculated using complex microphysical and optical algorithms (i.e., SAM and DDA) and those obtained from simplified assumptions as the layered-sphere models (even when the radial ice density distribution of the aggregated snowflakes is perfectly matched). Finally, some applications to quantitative precipitation estimation using radar data are presented to show how the resulting differences in the basic optical properties would propagate into radar measurable. Large discrepancies in the derivation of the equivalent water content and snowfall rate from radar measurements could be observed when large-size parameters are accounted for.

Description of the work

A new approach (SAM) that generates realistic dry and mixed-phase snow particles is developed. The study aims at simulating physically based complex snowflake habits that follow measured size-mass relations. The model uses simple hexagonal columns with variable linear dimensions and random orientation as building blocks. The implemented aggregation process is sufficiently versatile to model the aggregation of crystals of arbitrary shapes and sizes, but the growth of the snowflakes is statistically constrained to follow a specified size-mass relationship. The modeling of melting snowflakes allows to simulate the transition from the radiative properties of a dry snowflake to those of a mixed-phase particle which are of primary importance for the remote sensing of midlatitude precipitation. A direct comparison of the single-scattering properties of dry and wet snowflakes with the same mass, dimension, and shape is performed. Since a relevant melting process would change the morphology of the snowflake, the proposed method is not appropriate to model mixed-phase particles with large values of melted fraction.The radiative properties of the simulated particles are computed using the DDA method at 7 frequencies. Results from the ADDA-SAM models are compared to those of simplified spherical objects computed using the Mie solution. Different levels of accuracy for the representation of the radial mass distribution of the spherical targets are adopted. The highest degree of approximation of the complex aggregate using the spherical assumption is obtained by modeling a layered sphere with the exact radial density distribution of the SAM snowflake. The comparison between the radiative quantities calculated with the different models addresses the problem of the determination of the range of applicability of the spherical approximation for the modeling of complex snowflakes in radiative transfer studies. The analysis of the radiative properties obtained with the spherical models and the complex aggregated particles produced by SAM shows that the former are inadequate to represent the scattering characteristics of large aggregated particles. In particular, for size parameters larger than 3 the ADDA-SAM backscattering cross-section values are up to 3 orders of magnitude larger than those of spheres. Moreover, backscattering cross sections of the ADDA-SAM models do not show any strong resonance effect.Large differences in the phase functions computed with ADDA-SAM or assuming a Mie-spherical model are observed. Otherwise, differences in the total scattering cross-sections are less significant.The backscattering cross sections are integrated over a set of PSDs in order to evaluate the sensitivity of radar parameters to the different models of snowflakes. Substantial differences between the radiative properties calculated using complex models (such as ADDA-SAM) and those obtained from simplified spherical models (Mie applied to homogeneous sphere) are observed. However, when a Mie solution is applied to a layered-sphere model with same radial density distribution as the one obtained for the aggregates, differences in the Zhh are much less intense for the PSDs considered. Calculations of the radar equivalent reflectivity are sensitive to the radar backscattering model as well as the mass-dimension relationship and the PSDs used to describe the precipitation. For this reason it is fundamental to develop a robust characterization of ice and snow microphysical properties based on observations in order to properly model the radiative properties of snowflakes and develop robust Z-SR relationships.
Ori et al 2014
The aggregation scheme in four steps
Ori et al 2014
Image of a dry aggregate snowflake (left) and the same aggregate (right) with 10% of melted mass (red dots)
Ori et al 2014
Mass distribution within aggregates
Ori et al 2014
Backscattering cros section as a function of the size parameter
Ori et al 2014
Difference between wet and dry aggregate scattering properties