On Composition, Morphology, and Size Distribution of Airborne Mineral Dust - Mineral Dust: A Key Player in the Earth System - page 31

2.4 Size Distributions

The size distribution of an aerosol is one of its most important properties, as
it governs its impact on any process, in particular on its radiative impact - for
example, the scattering cross section scaling with powers of 3-5 of particle size -
its impact on multiphase chemical reactions (particle surface area/surface to volume
ratio) and on cloud processes (threshold sizes), but also its removal processes
efficiencies (sedimentation, wet removal) and as a consequence transport distances.
The mineral dust size distribution, which usually ranges from less than 100 nm
to more than 100 m, is notoriously difficult to measure. A major problem is
the collection efficiency of any instrument for particle sizes larger than a few
micrometers, particularly for airborne measurements. This can be in part overcome
by using inlet-free instruments like free-path optical-scattering instruments or body
impactor techniques for particle collection. Another problem arises from the sizing
technique itself, which usually uses a secondary measure - that is, light-scattering,
aerodynamic behavior - to determine particle size. While the relationship of these
secondary measures for spherical particles of known composition is generally well
known, this does not necessarily apply to mineral dust particles, which are highly
nonspherical and consist of mixtures of many different minerals in different shapes.
Geometric sizing techniques analyze samples collected by filtration or impactor
sampling by means of optical or electron microscopy (Fletcher et al. 2011 ). They
usually measure the two-dimensional projected shape of the particles (projected area
diameter). As a result, they are sensitive to particle orientation on the substrate and
overestimate particle size for flatly oriented, platelike particles (e.g., clay minerals).
However, this is the only major technique directly assessing the particle's geomet-
rical size. Aerodynamic and electrical methods determine the quantity of particle
with certain vacuum stopping distances and mechanical (super-micron range) or
electrical (submicron) mobilities. These measured quantities are then translated
into a geometrical size assuming spherical or spheroidal particles of homogeneous,
known density. Nonsphericity of the particles is assessed for aerodynamic methods
with a shape factor (e.g., like done by Kaaden et al. 2009 ), but still dust particles are
usually undersized (Reid et al. 2008 ). Light-scattering methods measure an optically
equivalent diameter with the same scattering properties. Thus, they are sensitive
to the particle shape and to its refractive index, imaginary as well as real part.
Assumption of sphericity usually leads to a slight oversizing (Collins et al. 2000 ).
Particle light absorption can strongly influence light-scattering sizing, especially for
forward scattering (Weinzierl et al. 2009 ; Schumann et al. 2011 ). Quality control
of size distribution measurements can be achieved by comparison with a reliable
integral measure, for example, mass concentration or optical extinction properties
(Collins et al. 2000 ; Esselborn et al. 2009 ; Schladitz et al. 2011 ).
A compilation of coarse mode modal diameters for desert aerosol in and
downwind of major deserts is compiled in Fig. 2.2 . In the Saharan Desert and
downwind, there is a gradient of decreasing modal diameters from ground-based
over airborne to medium-/long-range transported dust as it would be expected