Detection and characterization of birch pollen in the atmosphere using a multiwavelength Raman polarization lidar and Hirst-type pollen sampler in Finland

Stephanie Bohlmann,Sami Romakkaniemi,Mika Komppula,Maria Filioglou,Elina Giannakaki,Xiaoxia Shang,Annika Saarto

doi:10.5194/acp-19-14559-2019

Abstract

Abstract. We present the results of birch pollen characterization using lidar and in situ measurements based on a 11 d pollination period from 5 to 15 May 2016 at the European Aerosol Research Lidar Network (EARLINET) station in Vehmasmäki (Kuopio; 62∘44′ N, 27∘33′ E), Finland. The ground-based multiwavelength Raman polarization lidar PollyXT performed continuous measurements at this rural forest site and has been combined with a Hirst-type volumetric air sampler, which measured the pollen type and concentration at roof level (4 m). The period was separated into two parts due to different atmospheric conditions and detected pollen types. During the first period, high concentrations of birch pollen were measured with a maximum 2 h average pollen concentration of 3700 grains m−3. Other pollen types represented less than 3 % of the total pollen count. In observed pollen layers, the mean particle depolarization ratio at 532 nm was 10±6 % during the intense birch pollination period. Mean lidar ratios were found to be 45±7 and 55±16 sr at 355 and 532 nm, respectively. During the second period, birch pollen was still dominant, but a significant contribution of spruce pollen was observed as well. Spruce pollen grains are highly nonspherical, leading to a larger mean depolarization ratio of 26±7 % for the birch–spruce pollen mixture. Furthermore, higher lidar ratios were observed during this period with mean values of 60±3 and 62±10 sr at 355 and 532 nm, respectively. The presented study shows the potential of the particle depolarization ratio to track pollen grains in the atmosphere.

Highlights

Atmospheric pollen is a well-known health threat as it can irritate the respiratory system and cause asthmatic symptoms (Bousquet et al, 2008)
We present two case studies representative for different pollen mixtures: in the first case study only birch pollen had been detected by the Hirst-type sampler, and in the second case study spruce pollen was detected in addition to birch
355, 532 and 1064 nm; the particle extinction coefficient at 355 and 532 nm; the lidar ratio (LR) at 355 and 532 nm; the particle depolarization ratio (PDR) at 532 nm; the Ångström exponents calculated both from the backscatter coefficient at 355–532 nm and 532–1064 nm and from extinction coefficients at 355–532 nm; and the relative humidity from lidarderived water vapor profiles and temperature profiles from a radiosonde launched at 18:00 UTC

Summary

Introduction

Atmospheric pollen is a well-known health threat as it can irritate the respiratory system and cause asthmatic symptoms (Bousquet et al, 2008). 879 active stations have continuously monitored pollen type and concentration near ground level in 2016 (Buters et al, 2018). The majority of these stations operate with Hirst-type volumetric air samplers. Novel techniques have been developed to enable automated pollen monitoring and reduce workload Those techniques use, for example, automated image recognition (Oteros et al, 2015) or fluorescence spectra (Crouzy et al, 2016; Richardson et al, 2019; Saito et al, 2018) to identify pollen types, and they could enable a systematic pollen monitoring at ground level in near-real-time.

Objectives

Methods

Results

Conclusion