ice nucleation in mixed-phase clouds can already occur
at a temperature higher than - 20 °C (Seifert et al., 2010).
In contrast, laboratory studies showed that, e.g., mineral
dust and soot particles, which are two of the major
components of ice crystal residues (e.g., Pratt et
al., 2009), function as ice nuclei (IN) at lower
temperatures. One possible explanation for the observed
temperature differences might be the presence of
biological particles (one out of three particles have
biological origin, Pratt et al., 2009) acting as IN already
at temperatures above - 20 °C. Biological particles such
as bacteria and pollen possess membrane proteins and
non-proteinaceous macro-molecules, respectively, acting
as template for ice cluster formation reducing the energy
barrier required for ice nucleation. Nevertheless, the
dominating freezing mechanism, the properties inducing
ice nucleation, and the relative importance of biological
particles in the atmosphere are still unclear.
In our study, the immersion freezing behaviour of
different size-segregated biological particles is
investigated at the laminar flow tube LACIS (Leipzig
Aerosol Cloud Interaction Simulator, Hartmann et
al., 2011). SNOMAX®, outer membrane vesicles (OMV)
and surface material of pollen was used for the analysis.
SNOMAX® is considered as convenient surrogate for
bacterial IN and contains ice nucleation active proteins.
Bacterial strains, here extracted and cultured from rain
samples (Temkiv et al., 2011), have the ability to
produce outer membrane vesicles, which might also
contain ice nucleation active proteins. Surface material
washed off from pollen grains (e.g. birch) includes
macro-molecules which induce ice nucleation (Pummer
et al., 2011).
Figure 1 shows the ice fraction fice (number of
frozen droplets per sum of frozen and unfrozen droplets)
as function of temperature for all biological particles
investigated. For 800 nm SNOMAX® the freezing curve
is very steep in a temperature range from -5 °C to -10 °C
and levels off to constant values of about 0.4, i.e., 40 %
of SNOMAX® particles are ice active. The ice fractions
observed for the OMV containing particles where close
to the detection limit (in the order of 1 % and lower).
This indicates that only a small fraction of the OMV
particles nucleate ice. However, considering the small
size of the OMV particles (ranging between 50 nm and
160 nm as shown in Temkiv et al., 2011) it might be
possible that compared to SNOMAX®, the significantly
reduced IN ability is mainly due to size effects. The IN
contained in the washing water of birch pollen grains
nucleated ice below -18 °C with fice curve having a
slightly shallower slope than observed for SNOMAX®.
About 80 % of the 800 nm particles might include at
least one ice nucleation macro-molecule.
In conclusion, investigating the ice nucleation behavior
of quasi monodisperse biological IN, SNOMAX® was
found to be the most ice nucleation active substance,
followed by the residues of birch pollen washing water.
Outer membrane vesicles featured the lowest ice
nucleation potential.
This work is part of the DFG research unit INUIT,
under contract WE 4722/1-1.
Hartmann, S. et al. (2011) Atmos. Chem. Phys., 11,
1753-1767.
Pratt, K. et al. (2009) Nat. Geosci., 2, 397-400.
Pummer, B. G. et al. (2011) Atmos. Chem. Phys.
Discuss., 11, 27219-27241.
Seifert, P. et al. (2010) J. Geophys. Res.-Atmos., 115, 13.
Temkiv, T. S. et al. (2011) Bacteria in Clouds. PhD
dissertation, Graduate School of Science and
Technolopgy, Aarhus University, Denmark.
