Optical characterization of pure pollen types using a multi-wavelength Raman polarization lidar

Abstract. We present a novel algorithm for characterizing the
optical properties of pure pollen particles, based on the depolarization
ratio values obtained in lidar measurements. The algorithm was first tested
and validated through a simulator and then applied to the lidar
observations during a 4-month pollen campaign from May to August 2016 at
the European Aerosol Research Lidar Network (EARLINET) station in Kuopio
(62∘44′ N, 27∘33′ E), in Eastern Finland. With
a Burkard sampler, 20 types of pollen were observed and identified from concurrent measurements, with birch (Betula), pine (Pinus), spruce (Picea), and nettle (Urtica) pollen being the most
abundant, contributing more than 90 % of the total pollen load, regarding
number concentrations. Mean values of lidar-derived optical properties in
the pollen layer were retrieved for four intense pollination periods (IPPs).
Lidar ratios at both 355 and 532 nm ranged from 55 to 70 sr for all pollen
types, without significant wavelength dependence. An enhanced depolarization
ratio was found when there were pollen grains in the atmosphere, and an even
higher depolarization ratio (with mean values of 0.25 or 0.14) was observed
with the presence of the more non-spherical spruce or pine pollen. Under the
assumption that the backscatter-related Ångström exponent between 355
and 532 nm should be zero for pure pollen, the depolarization ratio of pure pollen particles at 532 nm was assessed, resulting in 0.24±0.01 and
0.36±0.01 for birch and pine pollen, respectively. Pollen optical
properties at 1064 and 355 nm were also estimated. The
backscatter-related Ångström exponent between 532 and 1064 nm was
assessed to be ∼0.8 (∼0.5) for pure birch (pine)
pollen; thus the longer wavelength would be a better choice to trace pollen in
the air. Pollen depolarization ratios of 0.17 and 0.30 at 355 nm were
found for birch and pine pollen, respectively. The depolarization values
show a wavelength dependence for pollen. This can be the key parameter for
pollen detection and characterization.


1 Particle linear depolarization ratio calculation (Eq.3 in the manuscript) We follow the detailed calculations in Tesche et al. 2009.Lidar-derived particle depolarization ratio ( particle ) can be expressed as the ratio of cross-( ⊥ ) and parallel-( ⫽ ) polarized particle backscatter coefficient: The particle backscatter coefficient  particle is the sum of cross-and parallel-polarized particle backscatter coefficient of both pollen and background aerosols: The depolarization ratio of one particle type can be defined as: The index x=pollen or background denotes the contribution of pollen or background particles, respectively.We can use the following relationships mathematically: We replace equations S5 and S6 in equation S2, the particle linear depolarization ratio can be then calculated using the particle backscatter coefficients ( pollen and  background ) and the depolarization ratios of both particle types ( pollen and  background ): (3) 2 Relationship of Å  and   (Eq.5 in the manuscript) Two aerosol populations, pollen (depolarizing) and background (non-depolarizing) aerosols are considered.The backscatter coefficient of the total particles is the sum of backscatter coefficient of both pollen and background aerosols: particle ( 1 ) =  pollen ( 1 ) +  background ( 1 ) (S7a) Similar as Eq.2 in the manuscript, the backscatter-related Ångström exponent (Å) can also be expressed in this equation: The index x=pollen, background or particle denotes the backscatter-related Ångström exponent of pollen, background or total particles.
We replace the top part of right side of Eq.S8 with x= particle with Eq.S7.And further use expression of  pollen ( 2 ) and  background ( 2 ) to replace the  pollen ( 1 ) and  background ( 1 ) in Eq.S7a, based on Eq.S8.Thus we have: After replacing  background ( 2 ) with Eq.S7b, the equation can be expressed as: Using the definition of pollen backscatter contribution (Eq.4 in the manuscript), a linear relationship between and  pollen ( 2 ) can be retrieved for the wavelength pair ( 1 ,  2 ): A similar formulate is found for the wavelength pair ( 2 ,  1 ) when considering  pollen ( 1 ):

Figure S 2 .
Figure S 1. Block diagram of the end-to-end simulator.Detail description is in section 3.

Figure S 3 .
Figure S 3. Estimated parameter ƞ  against the related assumed pollen depolarization ratio   at 532 nm.ƞ  is the ƞ(  ) value for the pure pollen (100 % pollen in the observed aerosol particle population,   = ), where ƞ is a parameter (Eq.6) using backscatter-related Ångström exponent between 355 and 532 nm (Å).Linear regression line is drawn by black dotted line, with fitting equation shown.The correlation coefficient (R 2 ) value is also given.The initial values (shown by the black pentagram) are 0.35 for pollen depolarization ratio and 0.25 for Å  (i.e.1.11 for ƞ  ).As we assume ƞ ̂ =1 (i.e.Å ̂ =0) in the inverse model for the retrieval, the final result of 0.39 for pure pollen is found (by the black triangle).

Figure S 4 .
Figure S 4. Examples of estimated bias and relative uncertainty on retrieved pollen depolarization ratio (DRpollen) against the initial values of backscatter-related Ångström exponent between 355 and 532 nm (Å) for pollen.The pollen Å is assumed as 0 (i.e.Å ̂ =0) in the inverse model for the calculation (shown by black dotted line).
Figure S 8. Mean values of the parameter  ƞ against pollen backscatter contribution at 355 nm   () inside the pollen layers during IPP-1 (a1-a4), or IPP-3 (b1-b4).The parameter  ƞ is a function (  ƞ = (   ) −Å  (,) in Eq.6) of the backscatter-related Ångström exponent between 355 and 532 nm for the total particle backscatter coefficients.The pollen depolarization ratio at 355 nm ( , ) is assumed to be 0.1, 0.2, 0.3, or 0.4 (from top to bottom).Linear regression lines are drawn by dotted lines, with fitting equations shown (Eq.5 or 8).The correlation coefficient (R 2 ) is also given.The size denotes the total pollen concentrations measured by the Burkard sampler on roof level; the colour represents the number concentration of the dominant pollen (birch for (a) and pine for (b)) against the total pollen number concentration.